A multi-dimensional weak link identification method and device for reliability improvement

Abstract

The present invention provides a multi-dimensional weak link identification method for reliability improvement, including determining the components and dominant scenarios of the weak links that need to be analyzed, obtaining component failure parameters and scenario parameters; determining the component failure impact type of each component failure on the load node, and calculating the reliability index of the distribution system under the influence of the component failure; taking partial derivatives of the time-invariant component failure parameters, calculating the sensitivity of the time-invariant component failure parameters under a single load level and in a source-load multi-period time sequence scenario, and identifying the reliability weak components under a single load level and in a source-load multi-period time sequence scenario; defining a time-varying scenario matrix, and according to the basic principles of the finite difference method, analytically calculating the sensitivity of the time-varying scenario parameters under a source-load multi-period time sequence scenario, and identifying the reliability dominant scenario under the source-load multi-period time sequence scenario. The present invention can provide a certain research basis for the planning, operation, and daily operation and maintenance of distribution systems based on reliability.

Description

Technical Field

[0001] The present invention belongs to the technical field of safety monitoring, and in particular relates to a multi-dimensional weak link identification method and device for improving reliability. Background Art

[0002] Identifying weak links in distribution system reliability relies on sensitivity analysis of various reliability-influencing factors and is a key application of distribution system reliability assessment in actual grid operations and maintenance. Currently, there are two main approaches for identifying weak links in power system reliability: finite difference methods, also known as perturbation methods, and direct derivation methods.

[0003] A typical finite difference method involves constructing a distribution network outage reliability prediction model based on a BP neural network, manually varying the influencing factor data over time, and conducting sensitivity analysis on each factor. A typical direct derivation method involves establishing the first- and second-order partial derivatives of the expected energy shortage (EENS) with respect to line capacity, ignoring the higher-order terms in the second-order Taylor series expansion, and conducting sensitivity analysis on the influencing factors.

[0004] Existing sensitivity analysis methods for reliability factors primarily focus on the impact of various component failure parameters, such as equipment failure rate and repair time, and switch operation time, on distribution system reliability. By repeatedly varying the values ​​of these failure parameters, the change in distribution system reliability indicators is calculated, and the impact of these parameters is then compared. However, these methods require a large number of repeated calculations of reliability indicators, resulting in low computational efficiency. Summary of the Invention

[0005] In view of this, the object of the present invention is to provide a multi-dimensional weak link identification method and device for reliability improvement, so as to solve the problem of low efficiency in calculating reliability indicators in the current weak link identification of distribution networks.

[0006] The present invention solves the technical problem by adopting the following technical solutions:

[0007] In a first aspect, a multi-dimensional weak link identification method for reliability improvement includes the following steps:

[0008] Identify the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters;

[0009] Among them, the component fault parameters include time-invariant component fault parameters under a single load level and time-invariant component fault parameters under a source-load multi-period timing scenario; the scenario parameters are time-varying scenario parameters under a source-load multi-period timing scenario.

[0010] Obtain power system parameters, determine the type of component failure impact on load nodes, and calculate the reliability index of the distribution system under the influence of component failure;

[0011] Based on the power transfer mode and power restoration time after the load node is out of power due to component failure, the four types of component failure impacts on the load node are determined;

[0012] Based on the definition of the impact types of four types of component failures, the four types of component failure impact correlation matrix are defined;

[0013] The distribution system reliability indicators include the average power outage duration SAIDI, the expected power shortage EENS, and the average power outage frequency SAIFI.

[0014] For the time-invariant component fault parameters under a single load level, partial derivatives of the time-invariant component fault parameters are obtained based on the obtained distribution system reliability index, and a sensitivity analysis formula is obtained. The sensitivity of the time-invariant component fault parameters under a single load level is calculated, and the reliability weak components under a single load level are identified.

[0015] Specifically, the sensitivity analysis topology of the time-invariant component fault parameters under a single load level is obtained;

[0016] Based on the obtained topology, the sensitivity of the system average power outage duration SAIDI, expected power shortage EENS, and average power outage number SAIFI to the failure rate of components in different locations is calculated. The sensitivity values ​​of the same parameter are normalized and compared with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis.

[0017] Based on the obtained topology, the sensitivity of the system average outage duration SAIDI and the expected power shortage EENS to the fault repair time of components at different locations, the section switch operation time, and the tie switch operation time are calculated. The sensitivity values ​​of the same parameter are normalized and compared with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis.

[0018] The time-invariant component fault parameters under a single load level include equipment failure rate, equipment failure repair time, section switch operation time, and tie switch operation time;

[0019] For the time-invariant component fault parameters in the source-load multi-period timing scenario, the partial derivative of the time-invariant component fault parameters is obtained based on the obtained distribution system reliability index, and the sensitivity analysis formula is obtained. The sensitivity of the time-invariant component fault parameters in the source-load multi-period timing scenario is analytically calculated, and the reliability weak components in the source-load multi-period timing scenario are identified; specifically,

[0020] Based on the calculated distribution system reliability index, the partial derivative of the component fault parameter is obtained to obtain the sensitivity formula of the reliability index affected by the time-invariant component fault parameter.

[0021] Calculate the sensitivity of the system average outage duration SAIDI, expected power shortage EENS, and system average outage frequency SAIFI to the failure rate of components in different locations. Homogenize the different data, retain the actual values, and then compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis.

[0022] Calculate the sensitivity of the system average outage duration SAIDI and expected power shortage EENS to the fault repair time of components at different locations, the operation time of sectionalizing switches, the operation time of tie switches, and the time of island formation. Homogenize the different data, retain the actual values, and then compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis.

[0023] Aiming at the time-varying scenario parameters in the source-load multi-period time series scenario, a time-varying scenario matrix is ​​defined. According to the basic principle of the finite difference method, the sensitivity of the time-varying scenario parameters in the source-load multi-period time series scenario is analytically calculated, and the sensitivity analysis formula is obtained to identify the reliability-dominated scenario in the source-load multi-period time series scenario.

[0024] In a second aspect, a multi-dimensional weak link identification device for reliability improvement includes:

[0025] A parameter acquisition module is used to determine the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters; the component failure parameters include time-invariant component failure parameters under a single load level and time-invariant component failure parameters under a source-load multi-period time sequence scenario; the scenario parameters are time-varying scenario parameters under a source-load multi-period time sequence scenario;

[0026] A reliability index acquisition module is used to obtain power system parameters, determine the type of component failure impact of each component failure on the load node, and calculate the reliability index of the distribution system under the influence of the component failure;

[0027] A module for identifying reliability weak components under a single load level is used to determine the partial derivative of the time-invariant component fault parameters under a single load level based on the obtained distribution system reliability index, obtain a sensitivity analysis formula, calculate the sensitivity of the time-invariant component fault parameters under a single load level, and identify reliability weak components under a single load level. The time-invariant component fault parameters under a single load level include equipment failure rate, equipment fault repair time, section switch operation time, and tie switch operation time.

[0028] A module for identifying weak reliability components in a source-load multi-period timing scenario is used to determine the partial derivative of the time-invariant component fault parameters in the source-load multi-period timing scenario based on the obtained distribution system reliability index, analytically calculate the sensitivity of the time-invariant component fault parameters in the source-load multi-period timing scenario, and identify weak reliability components in the source-load multi-period timing scenario. The time-invariant component fault parameters in the source-load multi-period timing scenario include equipment failure rate, equipment fault repair time, section switch operation time, tie switch operation time, and island formation time.

[0029] The reliability-dominated scenario identification module in the source-load multi-period timing scenario is used to define a time-varying scenario matrix for the time-varying scenario parameters in the source-load multi-period timing scenario, and analyze and calculate the sensitivity of the time-varying scenario parameters in the source-load multi-period timing scenario based on the basic principles of the finite difference method, so as to identify the reliability-dominated scenario in the source-load multi-period timing scenario; the time-varying scenario parameters include load node demand, distributed power output and energy storage element output.

[0030] The embodiments of the present invention bring the following beneficial effects:

[0031] The multidimensional weak link identification method for reliability improvement studied in this invention uses a direct derivation method to conduct sensitivity analysis on the time-invariant component fault parameters to identify weak components that restrict reliability improvement. A time-varying scenario matrix is ​​established for time-varying scenario parameters, and a finite difference method is used to conduct sensitivity analysis to identify the dominant scenarios that affect the reliability level. The multidimensional weak link identification method for reliability improvement proposed in this invention can provide a certain research foundation for the planning, operation, and daily maintenance of distribution systems based on reliability.

[0032] Other features and advantages of the present invention will be described in the following description, and some features will become obvious from the description or be understood by practicing the present invention. The objectives and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of the structure of a multi-dimensional weak link identification method for reliability improvement provided by an embodiment of the present invention;

[0035] Figure 2 A schematic diagram of a topological structure for sensitivity analysis of fault parameters of time-invariant components under a single load level provided by an embodiment of the present invention;

[0036] FIG3( a ) is a schematic diagram showing the equipment failure rate sensitivity results of time-invariant component failure parameters under a single load level provided by an embodiment of the present invention;

[0037] FIG3( b ) is a schematic diagram showing the equipment fault repair time sensitivity results of time-invariant component fault parameters under a single load level provided by an embodiment of the present invention;

[0038] FIG4( a ) is a schematic diagram showing the sensitivity of the average number of power outages in the system to the failure rates of components at different locations in a multi-period source-load time sequence scenario provided by an embodiment of the present invention;

[0039] FIG4( b ) is a schematic diagram showing the sensitivity of the average power outage time of the system to the failure rate of components at different locations in a scenario taking into account the source-load multi-period time sequence according to an embodiment of the present invention;

[0040] FIG4( c ) is a schematic diagram showing the sensitivity of the expected power shortage to the failure rate of components at different locations in a multi-period source-load time sequence scenario provided by an embodiment of the present invention;

[0041] FIG5(a) is a schematic diagram showing the sensitivity of the average power outage time of the system to the repair time of component equipment at different locations in a multi-period source-load time sequence scenario provided by an embodiment of the present invention;

[0042] FIG5( b ) is a schematic diagram showing the sensitivity of the expected power shortage to the fault repair time of component equipment at different locations in a multi-period source-load time sequence scenario provided by an embodiment of the present invention;

[0043] Figure 6 A schematic diagram of the sensitivity results of time-varying scenario parameters provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] The present invention proposes a multi-dimensional weak link identification method and device for reliability improvement. In view of the fact that a large number of distributed power sources, energy storage devices and flexible loads are connected to the new power distribution system, the scenario parameters such as the output of distributed power sources, energy storage devices and load requirements in various scenarios also affect the fault recovery capability of the power distribution system, and thus affect the reliability level of the power distribution system. Further research is conducted on the sensitivity analysis method of component fault parameters to form the analysis results of weak components that restrict reliability improvement. The specific introduction is as follows:

[0046] Example 1

[0047] This embodiment provides a multi-dimensional weak link identification method for reliability improvement. Figure 1 As shown, the following steps are included:

[0048] S1. Determine the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters. The component failure parameters include time-invariant component failure parameters under a single load level and time-invariant component failure parameters under a source-load multi-period sequence scenario; the scenario parameters are time-varying scenario parameters under a source-load multi-period sequence scenario;

[0049] S2. Obtain power system parameters, determine the impact type of each component failure on the load node, calculate the power outage time and other distribution system reliability indicators of each load point under the influence of the component failure, and define the component failure impact correlation matrix based on the component failure impact type. Distribution system reliability indicators include the system average power outage duration SAIDI, expected power shortage EENS, and the system average power outage number SAIFI. The specific steps are:

[0050] S2.1. Based on the power supply transfer method and power restoration time after the load node is out of power due to component failure, determine the four types of component failure impacts on the load node. Including:

[0051] Type a: A branch fault causes all power supply paths to the load node to be interrupted. The load node needs to wait until the fault is repaired before power can be restored. The power outage duration caused by this branch fault is the branch fault repair time.

[0052] Type b: A branch fault causes all power supply paths to the load node to be interrupted. The load node needs to wait for the fault to be isolated and then restored by the main power supply. The power outage duration caused by this branch fault is the fault isolation time.

[0053] Type c: A branch fault causes all power supply paths to the load node to be interrupted. The load node needs to wait for the fault to be isolated and then restored by the backup power supply. The power outage duration caused by this branch fault is the time required to isolate the fault and restore power.

[0054] Type d: The branch fault has no impact on the load node, for example, the faulted branch and node are on different feeders.

[0055] S2.2. Based on the definition of the four types of component failure impact types, the four types of component failure impact association matrices (FEIM) are defined as FIEMA, FIEMB, FIEMC, and FIEMD. The definitions are as follows:

[0056]

[0057] Define the fault check matrix CHECK as:

[0058] CHECK=FEIM A+FEIM B+FEIM C+FEIM D (5)

[0059] Among them, under the premise of accurate identification of the fault impact type, the elements in the fault check matrix CHECK should all be 1.

[0060] S3. For the time-invariant component fault parameters under a single load level, find the partial derivative of the time-invariant component fault parameters based on the obtained distribution system reliability index, analyze and calculate the sensitivity of the time-invariant component fault parameters under a single load level, identify the reliability weak components under a single load level, and obtain the weak link identification results. Figure 2 As shown, specifically,

[0061] S3.1. Obtain the topology of the distribution network and the sensitivity analysis topology of the time-invariant component failure parameters under a single load level, wherein the time-invariant component failure parameters under a single load level include the equipment failure rate λ i , Equipment failure repair time μ i , Section switch operation time t sw , contact switch operation time;

[0062] S3.2. Based on the obtained topology, calculate the sensitivity of the system average outage duration SAIDI, expected power shortage EENS, and average number of system outages SAIFI to the failure rates of components in different locations. Normalize the sensitivity values ​​of the same parameter and compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis;

[0063] The equipment failure rate SAIDI-λ under a single load level i Taking the sensitivity calculation method as an example, first find the partial derivative as follows:

[0064]

[0065] Where λ i represents the equipment failure rate of the i-th component; n j Indicates the number of users of the jth load node; column (A i )、column(B i ) and column(C i ) represent the i-th column of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC respectively; row(A i )、row(B i) and row(C i ) represents the i-th row of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC respectively.

[0066] Repeat the above partial derivative calculation process to obtain the reliability level equipment failure rate λ under a single load level i The sensitivity analysis formula affected by each fault parameter is shown in Table 1 below:

[0067]

[0068] Since the sensitivity comparison between different parameters is meaningless, the sensitivity values ​​of the same parameter are homogenized. Combining Figure 3(a) and Table 1, the failure rate of the weak link equipment λ is obtained. i The identification results and cause analysis are as follows:

[0069] 1) Weak link identification results

[0070] Figure 3(a) is a schematic diagram of the equipment failure rate sensitivity results of the time-invariant component failure parameters under a single load level, with SAIFI_λ i Sensitivity analysis results, SAIDI_λ i Sensitivity analysis results and EENS_λ i Sensitivity analysis results.

[0071] From SAIFI_λ i The sensitivity analysis results show that the failure rate of components at different positions in the same feeder is λ i The impact on SAIFI is the same. i The location with the greatest sensitivity occurs on the feeder where nodes 17-38 are located;

[0072] From SAIDI_λ i The sensitivity analysis results show that the failure rate of components at different positions in the same feeder is λ i The degree of impact on SAIDI is also different. i The top five most sensitive positions are 45, 40, 46, 39, and 42;

[0073] From EENS_λ i The sensitivity analysis results show that the failure rate of components at different positions in the same feeder is λ i The degree of impact on EENS is also different. i The top five most sensitive positions are 45, 40, 46, 39 and 105.

[0074] 2) Analysis of causes of weak links

[0075] According to SAIFI_λ in Table 1 i Sensitivity analysis calculation formula, SAIFI_λ i The sensitivity of is only related to the sum of the i-th row elements of the fault impact correlation matrix FIIMA, FIIMB and FIIMC and the number of users on each node. The specific type of fault has no effect on the sum of the i-th row elements of the fault impact correlation matrix FIIMA, FIIMB and FIIMC, so SAIFI_λ i The sensitivity is only related to the total number of users on each feeder. i The sensitivity ranking is consistent with the ranking of the total number of users on each feeder, which is consistent with the actual situation.

[0076] According to SAIDI_λ in Table 1 i Sensitivity calculation formula, SAIDI_λ i The sensitivity of is related to the sum of the product of the power outage time of a load node j caused by the i-th component failure and the total number of users on node j. Different fault types cause different load power outage times. It is necessary to calculate the power outage time based on the values ​​of the i-th row elements of the fault impact correlation matrix FIEMA, FIEMB, and FIEMC, and accumulate them in combination with the total number of users on each node. i The sensitivity is ranked in the same order as the sum of the products, which is consistent with the actual situation.

[0077] According to EENS_λ in Table 1 i Sensitivity calculation formula, EENS_λ i The sensitivity of is related to the sum of the product of the outage time of a load node j caused by the i-th component failure and the load demand at node j. Different fault types cause different load outage times. It is necessary to calculate the outage time based on the values ​​of the i-th row elements of the fault impact correlation matrix FIEMA, FIEMB, and FIEMC, and then accumulate them in combination with the load demand of each node. EENS_λ in Figure 3(a) i The sensitivity is ranked in the same order as the sum of the products, which is consistent with the actual situation.

[0078] S3.3. Based on the obtained topology, calculate the sensitivity of the system average outage duration (SAIDI) and the expected power shortage (EENS) to the repair time of component equipment failures at different locations. Normalize the sensitivity values ​​of the same parameter and compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis.

[0079] Using the direct derivation method, the equipment failure repair time μ under a single load level is obtained i The sensitivity formula affected by each fault parameter is shown in Table 2.

[0080]

[0081] Since the sensitivity of different parameters is meaningless, the sensitivity values ​​of the same parameter are normalized. Combined with Figure 3(b) and Table 2, the weak link fault repair time μ is obtained. i The identification results and cause analysis are as follows:

[0082] 1) Weak link identification results

[0083] Figure 3(b) is a schematic diagram of the equipment fault repair time sensitivity results of the time-invariant component fault parameters under a single load level, with SAIFI_μ i Sensitivity analysis results, SAIDI_μ i Sensitivity analysis results and EENS_μ i Sensitivity analysis results.

[0084] From SAIDI_μ i The sensitivity analysis results show that there are some positions where SAIDI_μ i Sensitivity is 0. SAIDI_μ i The top five most sensitive positions are 122, 39, 125, 45 and 58.

[0085] From EENS_μ i The sensitivity analysis results show that there are some locations where EENS_μ i Sensitivity is 0, and these positions are related to EENS_μ i The position where sensitivity is 0 is exactly the same. i The top five most sensitive positions are 122, 39, 125, 45 and 58.

[0086] 2) Analysis of causes of weak links

[0087] According to SAIDI_μ in Table 2 i Sensitivity analysis calculation formula, SAIDI_μ i The sensitivity is related to the sum of the number of type A faults at a load node j caused by the i-th component and the total number of users at node j. i The sensitivity is consistent with the order of the sum of the products, which is in line with the actual situation. The power outage time caused by only type a fault is the equipment fault repair time μ i , SAIDI_μ i Faults at locations with a sensitivity of 0 will not cause a type of impact on all nodes in the system, so the SAIDI_μ at these locations i The sensitivity is 0.

[0088] According to EENS_μ in Table 2 i Sensitivity calculation formula, EENS_μ i The sensitivity is related to the sum of the number of type A faults at a load node j caused by the i-th component and the product of the load demand at node j. i The sensitivity is consistent with the order of the sum of the products, which is in line with the actual situation. The node power shortage ens caused by only type a fault and the equipment fault repair time μ i Related, EENS_μ i Locations with a sensitivity of 0 will not cause a type of impact on all nodes in the system, so the EENS_μ of these locations is i The sensitivity is 0.

[0089] S3.4. Based on the obtained topology, calculate the system average power outage time SAIDI and the expected power shortage EENS for the segmented switch operation time t of components at different positions. sw The sensitivity of the weak link segment switching operation time t is obtained sw Identify results and cause analysis;

[0090] Furthermore, the direct derivation method is used to obtain the section switch operation time t under a single load level. sw The calculation formula for sensitivity analysis affected by various fault parameters is shown in Table 3:

[0091]

[0092] S3.5. Based on the obtained topology, calculate the average power outage time SAIDI and the expected power shortage EENS for the failure rate of components at different locations λ i The sensitivity of the weak link equipment failure rate λ is obtained i Identify results and analyze causes;

[0093] Furthermore, the direct derivation method is used to obtain the tie switch operation time t under a single load level. op The sensitivity analysis calculation formula affected by each fault parameter is shown in Table 4 below:

[0094]

[0095] S4. For the time-invariant component fault parameters in the source-load multi-period sequence scenario, the partial derivative of the time-invariant component fault parameters is obtained based on the obtained distribution system reliability index, the sensitivity of the time-invariant component fault parameters in the source-load multi-period sequence scenario is calculated analytically, and the reliability weak components in the source-load multi-period sequence scenario are identified. The time-invariant component fault parameters in the source-load multi-period sequence scenario include the equipment failure rate λ i, Equipment failure repair time μ i , Section switch operation time t sw , contact switch operation time t op , island formation time t isld Specifically,

[0096] S4.1. Based on the calculated distribution system reliability index, find the partial derivative of the component fault parameter, that is:

[0097]

[0098] Where β represents the fault parameter of a time-invariant component in the distribution system, E(F) represents the expected value of a reliability index of the distribution system, and f k represents the reliability index corresponding to the kth typical scenario, P k represents the probability of the kth typical scenario occurring, and K represents the number of typical running scenarios after reduction.

[0099] S4.2. Calculate the average number of system outages SAIFI, the average system outage duration SAIDI, and the expected power shortage EENS for the failure rate λ of components at different locations i The sensitivity of the weak link equipment failure rate λ is obtained i Identification results and cause analysis.

[0100] Considering the equipment failure rate λ in the multi-period source-load time sequence scenario i Taking the sensitivity calculation method as an example, first find the partial derivative as follows:

[0101]

[0102] Where, P k represents the probability of the kth scenario occurring, SAIDI(k) represents the average system outage time of the kth scenario, and IRM(k) represents the island recovery matrix of the kth scenario.

[0103] Repeat the above partial derivative calculation process to obtain the equipment failure rate λ when considering the source-load multi-period time series scenario i The sensitivity analysis formula affected by the fault parameters of each time-invariant component is shown in Table 5 below:

[0104]

[0105] Equipment failure repair time μ in source-load multi-period time series scenario i The sensitivity results affected by the fault parameters of each time-invariant component are shown in Figure 4(a), Figure 4(b), and Figure 4(c). Compared with the calculation results in Table 5, since the maximum sensitivity values ​​in the two cases are different, the data are not normalized and the actual values ​​are retained.

[0106] Combining Figure 4(a) and Table 5, we get SAIFI_λ i The weak component identification results and cause analysis are as follows.

[0107] 1) Weak component identification results

[0108] As can be seen from Figure 4(a), regardless of whether DG or energy storage is connected, SAIFI_λ at any position i The sensitivity is exactly the same. i The location with the greatest sensitivity appears on the feeder where nodes 17-38 are located.

[0109] 2) Analysis of causes of weak components

[0110] According to SAIFI_λ in Table 5 i Sensitivity calculation formula, SAIFI_λ i The sensitivity of is only related to the sum of the i-th row elements of the fault impact correlation matrix FIEMA, FIEMB and FIEMC and the number of users at each node. The presence of DG and energy storage has no effect on the sum of the i-th row elements of the fault impact correlation matrix FIEMA, FIEMB and FIEMC, so SAIFI_λ i The sensitivity is only related to the total number of users on each feeder. i The sensitivity ranking is consistent with the ranking of the total number of users on each feeder, which is consistent with the actual situation.

[0111] Combining Figure 4(b) and Table 5, we get SAIDI_λ i The weak component identification results and cause analysis are as follows:

[0112] 1) Weak component identification results

[0113] For the position where DG and energy storage are not connected on the feeder, SAIDI_λ before and after connecting DG and energy storage i The results of sensitivity analysis were completely consistent.

[0114] For the location where DG and energy storage are connected on the feeder, SAIDI_λ i Sensitivity decreased, but the degree of decrease varied among different locations.

[0115] Regardless of whether DG or energy storage is connected, SAIDI_λ i The position with the highest sensitivity is 45.

[0116] 2) Analysis of causes of weak components

[0117] As can be seen from Figure 4(b), for the locations on the feeder line where DG and energy storage are not connected, the i-th row vector of the corresponding island recovery matrix IRM is always 0, and the power outage time index of each load node on these feeders remains unchanged. Therefore, the SAIDI_λ before and after the DG and energy storage are connected at these locations is i Same sensitivity.

[0118] For the locations where DG and energy storage are connected on the feeder, there is a differentiated potential to restore the load nodes originally affected by the type A fault, depending on the actual output in different scenarios. The power outage time of these load nodes will be reduced from the fault repair time μ i Reduced to the island formation time tisld, correspondingly, the SAIDI_λ of these load nodes i Sensitivity also depends on the fault repair time μ i The product of the time it takes to form an island and the number of users at the load node is reduced to the product of the time it takes to form an island and the number of users at the load node. The extent of the reduction is affected by the island recovery capabilities of DG and energy storage in different scenarios. For locations that are not within the island recovery range of DG and energy storage, their SAIDI_λ i The sensitivity remains unchanged and is consistent with the actual situation.

[0119] Combining Figure 4(c) and Table 5, we get EENS_λ i The weak component identification results and cause analysis are as follows:

[0120] 1) Weak component identification results

[0121] As can be seen from Figure 4(c), after connecting DG and energy storage, the EENS_λ at any position i The sensitivity is reduced. i The position with the highest sensitivity is 45.

[0122] 2) Analysis of causes of weak components

[0123] Considering the time series fluctuation characteristics of load, the load demand of all nodes in each typical scenario is different, rather than taking a constant maximum load as the load demand of each load node. At the same time, for feeders connected to DG and energy storage, there is a differentiated potential to restore the load nodes originally affected by Class A faults according to the actual output in different scenarios. The power outage time of these load nodes will be reduced from the fault repair time μ i Reduced to the island formation time tisld, accordingly, the EENS_λ of these load nodes i Sensitivity also depends on the fault repair time μ i The product of the load demand is reduced to the product of the island formation time tisld and the load demand. The extent of the reduction is affected by the island recovery capabilities of DG and energy storage in different scenarios, which is in line with the actual situation.

[0124] S4.3. Calculate the average system outage time SAIDI and the expected power shortage EENS for the repair time μ of the fault of the components at different locations i The sensitivity of the weak link equipment failure repair time μ is obtained i Identify results and analyze causes;

[0125] Using the direct derivation method, the equipment failure repair time μ is obtained when considering the source-load multi-period time series scenario. i The sensitivity analysis formula affected by the fault parameters of each time-invariant component is shown in Table 6.

[0126]

[0127] Equipment failure repair time μ when considering the source-load multi-period time sequence scenario i The sensitivity results affected by the fault parameters of each time-invariant component are shown in Figure 5(a) and Figure 5(b). Compared with the calculated results in Table 6, since the maximum sensitivity values ​​in the two cases are different, the data are not normalized and the actual values ​​are retained.

[0128] Combining Figure 5(a) and Table 6, SAIDI_μ i The weak component identification results and cause analysis are as follows:

[0129] 1) Weak component identification results

[0130] Comparison of SAIDI_μ in two cases in Figure 5(a) i Sensitivity analysis results show that there are some positions where SAIDI_μ i The sensitivity is 0.

[0131] For locations on the feeder line where there is no DG or energy storage connected, SAIDI_μ i Sensitivity was the same in both cases.

[0132] For the location where DG and energy storage are connected on the feeder, SAIDI_μ i Sensitivity decreased, but the degree of decrease varied among different locations.

[0133] SAIDI_μ i The position with the highest sensitivity occurs at 122.

[0134] (2) Analysis of causes of weak components

[0135] The power outage time caused by only type a fault is the equipment fault repair time μ i , while SAIDI_μ iThe locations with a sensitivity of 0 will not cause type A impact on all nodes in the system, so these locations have SAIDI_μ in both cases. i The sensitivity is 0.

[0136] For the locations on the feeder line that are not connected to DG or energy storage, the i-th row vector of the corresponding island recovery matrix IRM is always 0, and the power outage time index of each load node on these feeders remains unchanged. Therefore, these locations have SAIDI_μ in both cases. i The sensitivity is the same.

[0137] For the locations where DG and energy storage are connected on the feeder, there is a differentiated potential to restore the load nodes originally affected by Class A faults, depending on the actual output in different scenarios. The number of power outages at these load nodes affected by Class A faults will be reduced from the equipment failure rate λ i Reduced to 0, accordingly, the SAIDI_μ of these load nodes i Sensitivity also depends on the device failure rate λ i The product of the number of users is reduced to 0. The extent of the reduction is affected by the island recovery capabilities of DG and energy storage in different scenarios, which is consistent with the actual situation.

[0138] Combined with Figure 5(b) and Table 6, EENS_μ i The weak component identification results and cause analysis are as follows:

[0139] 1) Weak component identification results

[0140] Comparison of EENS_μ in two cases in Figure 5(b) i Sensitivity analysis results show that there are some positions where EENS_μ i The sensitivity is 0.

[0141] For the location where DG and energy storage are connected on the feeder, EENS_μ i Sensitivity decreased, but the degree of decrease varied among different locations.

[0142] EENS_μ i The position with the highest sensitivity occurs at 122.

[0143] 2) Analysis of causes of weak components

[0144] The power outage time caused by only type a fault is the equipment fault repair time μ i , and EENS_μ i Positions with a sensitivity of 0 will not cause type A impact on all nodes in the system, so these positions have EENS_μ in both cases. i The sensitivity is 0.

[0145] For the locations where DG and energy storage are connected on the feeder, there is a differentiated potential to restore the load nodes originally affected by Class A faults, depending on the actual output in different scenarios. The number of power outages at these load nodes affected by Class A faults will be reduced from the equipment failure rate λ i Reduced to 0, accordingly, the EENS_μ of these load nodes i Sensitivity also depends on the device failure rate λ i The product of the load demand is reduced to 0. The extent of the reduction is affected by the island recovery capabilities of DG and energy storage in different scenarios, which is in line with the actual situation.

[0146] S4.4. Calculate the average system outage time SAIDI and the expected power shortage EENS for the segmented switch operation time t of components in different positions sw sensitivity;

[0147] The direct derivation method is used to obtain the time-segmented switch operation time t considering the multi-period source-load time sequence scenario. sw The sensitivity analysis calculation formula affected by the fault parameters of each time-invariant component is shown in Table 7.

[0148]

[0149] S4.5. Calculate the average system outage time SAIDI and the expected power shortage EENS for the operating time t of the tie switch of different position components. op sensitivity;

[0150] The direct derivation method is used to obtain the tie switch operation time t when considering the multi-period source-load time sequence scenario. op The sensitivity analysis calculation formula affected by the fault parameters of each time-invariant component is shown in Table 8 below.

[0151]

[0152] S4.6. Calculate the average system outage time SAIDI and the expected power shortage EENS for the islanding time t of components at different locations. isld sensitivity;

[0153] The direct derivation method is used to obtain the island formation time t when considering the multi-period time series scenario of source and load isld The sensitivity analysis calculation formula affected by the fault parameters of each time-invariant component is shown in Table 9 below.

[0154]

[0155] S5. For the time-varying scenario parameters in the source-load multi-period timing scenario, define a time-varying scenario matrix, and based on the basic principles of the finite difference method, analyze and calculate the sensitivity of the time-varying scenario parameters in the source-load multi-period timing scenario, and perform reliability-dominated scenario identification in the source-load multi-period timing scenario; the time-varying scenario parameters include load node demand, distributed power output, and energy storage element output.

[0156] Time-varying scene matrix It is defined as,

[0157]

[0158] Where, Indicates the value of the i-th scenario parameter at the j-th node in the k-th scenario, p represents the number of types of time-varying scenario parameters, and N represents the number of nodes in the distribution system. Taking the three types of scenario parameters, load node demand, distributed power output, and energy storage element output, as examples, their sensitivity is analyzed. The time-varying scenario matrix at this time is You can write:

[0159]

[0160] Where, L k,j , DG k,j and ESS k,j They represent the load demand, output of connected distributed power generation and output of energy storage components of the jth node in the kth scenario respectively. The SAIFI(k) index is the same in different scenarios. Only the SAIDI(k) and EENS(k) indexes will change with the change of the island recovery matrix IRM(k). The SAIDI(k) and EENS(k) indexes are related to the time-varying scenario matrix. The functional form of can be abstractly expressed as shown in formulas (11) and (12).

[0161] SAIDI(k)=f1(IRM(k))=f1[g(TVS k )] (11)

[0162] EENS(k)=f2(IRM(k))=f2[g(TVS k )] (12)

[0163] Wherein, mapping symbols f1 and f2 represent the association between SAIDI(k) and IRM(k), EENS(k) and IRM(k), respectively, and mapping symbol g represents the association between IRM(k) and TVS obtained by k The relationship between them.

[0164] According to the basic principle of the finite difference method, the SAIDI(k)-TVS sensitivity can be calculated as follows:

[0165]

[0166] EENS(k)-TVS sensitivity can be calculated as follows:

[0167]

[0168] In the formula, the row vector Represents the average annual power outage times of N nodes, row vector Represents the annual average power outage time of N nodes, the row vector Represents the power shortage of N nodes, the row vector P 1×K =[P1,P2,…,P K ] represents the probability of K scenes appearing, μ(i) j LP ens(i) j LP They represent the power outage time and power shortage of the jth node in the i-th scenario respectively.

[0169] SAIFI(k), SAIDI(k) and EENS(k) represent the average number of power outages, average power outage time and expected power shortage in the kth scenario, respectively. n represents the row vector composed of the number of users of N load nodes arranged in numerical order. Ncons represents the total number of users in the distribution system. [row k (L)] T represents the transposition of the row vector formed by all load demands in the kth scenario, ||ens(k) LP ||1 means taking a vector The 1-norm of .

[0170] Scenario parameters such as load node demand, distributed generation output, and energy storage element output from eight typical scenarios were input into a time-varying scenario matrix. This generated a quantitative analysis of each scenario's contribution to the outage indicator from the overall perspective of the scenario parameters. A time-varying scenario parameter sensitivity analysis method calculated the sensitivity of the system's average outage duration (SAIDI) and expected power shortage (EENS) to the time-varying scenario parameters under different typical scenarios, thereby identifying the dominant scenarios affecting reliability.

[0171] In order to compare the contribution of each scenario to the power outage index, the sensitivity changes of time-varying scenario parameters under 8 typical scenarios are calculated. The results are as follows: Figure 6 As shown,

[0172] 1) Dominant scene identification results

[0173] Observing the two curves in the figure, the SAIDI(k)-TVS sensitivity and EENS(k)-TVS sensitivity in each scenario are both negative, that is, compared with the baseline scenario, the SAIDI(k) and EENS(k) in each scenario are reduced.

[0174] Judging from the absolute values, the scenarios with the greatest sensitivity to SAIDI(k)-TVS and EENS(k)-TVS appear in scenarios 4 and 6, respectively. That is, scenario 4 is the dominant scenario affecting the SAIDI(k) indicator, and scenario 6 is the dominant scenario affecting the EENS(k) indicator.

[0175] 2) Analysis of the causes of dominant scenarios

[0176] Due to the access of distributed power sources, energy storage and flexible loads, the load recovery capacity of the distribution system after failure has been improved, and the power outage time and power shortage indicators in each scenario have been reduced.

[0177] Among all typical scenarios, the distributed power output in Scenario 4 and Scenario 6 is low, and the load range restored under fault scenarios is small. Therefore, compared with the baseline scenario, their reliability improvement effect is limited, and they are the dominant scenarios affecting the reliability level.

[0178] Example 2

[0179] A multi-dimensional weak link identification device for reliability improvement, comprising:

[0180] A parameter acquisition module is used to determine the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters; the component failure parameters include time-invariant component failure parameters under a single load level and time-invariant component failure parameters under a source-load multi-period time sequence scenario; the scenario parameters are time-varying scenario parameters under a source-load multi-period time sequence scenario;

[0181] A reliability index acquisition module is used to obtain power system parameters, determine the type of component failure impact of each component failure on the load node, and calculate the reliability index of the distribution system under the influence of the component failure;

[0182] A module for identifying reliability weak components under a single load level is used to determine the partial derivative of the time-invariant component fault parameters under a single load level based on the obtained distribution system reliability index, obtain a sensitivity analysis formula, calculate the sensitivity of the time-invariant component fault parameters under a single load level, and identify reliability weak components under a single load level. The time-invariant component fault parameters under a single load level include equipment failure rate, equipment fault repair time, section switch operation time, and tie switch operation time.

[0183] A module for identifying weak reliability components in a source-load multi-period timing scenario is used to determine the partial derivative of the time-invariant component fault parameters in the source-load multi-period timing scenario based on the obtained distribution system reliability index, analytically calculate the sensitivity of the time-invariant component fault parameters in the source-load multi-period timing scenario, and identify weak reliability components in the source-load multi-period timing scenario. The time-invariant component fault parameters in the source-load multi-period timing scenario include equipment failure rate, equipment fault repair time, section switch operation time, tie switch operation time, and island formation time.

[0184] The reliability-dominated scenario identification module in the source-load multi-period timing scenario is used to define a time-varying scenario matrix for the time-varying scenario parameters in the source-load multi-period timing scenario, and analyze and calculate the sensitivity of the time-varying scenario parameters in the source-load multi-period timing scenario based on the basic principles of the finite difference method, so as to identify the reliability-dominated scenario in the source-load multi-period timing scenario; the time-varying scenario parameters include load node demand, distributed power output and energy storage element output.

[0185] It should be noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements that are inherent to such process, method, article or apparatus.

[0186] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and alterations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A multi-dimensional weak link identification method for reliability improvement, characterized in that: The following steps are included: Identify the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters; The component fault parameters include time-invariant component fault parameters under a single load level and time-invariant component fault parameters under a source-load multi-period sequence scenario; the scenario parameters are time-varying scenario parameters under a source-load multi-period sequence scenario; Obtain power system parameters, determine the type of component failure impact on load nodes, and calculate the reliability index of the distribution system under the influence of component failure; For the time-invariant component fault parameters under a single load level, partial derivatives of the time-invariant component fault parameters are obtained based on the obtained distribution system reliability index, and a sensitivity analysis formula is obtained. The sensitivity of the time-invariant component fault parameters under a single load level is calculated, and the reliability weak components under a single load level are identified. The time-invariant component fault parameters under a single load level include equipment failure rate, equipment failure repair time, section switch operation time, and tie switch operation time; For the time-invariant component fault parameters in the source-load multi-period timing scenario, the partial derivative of the time-invariant component fault parameters is calculated based on the obtained distribution system reliability index, and the sensitivity analysis formula is obtained. The sensitivity of the time-invariant component fault parameters in the source-load multi-period timing scenario is analytically calculated, and the reliability weak components in the source-load multi-period timing scenario are identified; Aiming at the time-varying scenario parameters in the source-load multi-period time series scenario, a time-varying scenario matrix is ​​defined. According to the basic principle of the finite difference method, the sensitivity of the time-varying scenario parameters in the source-load multi-period time series scenario is analytically calculated, and the sensitivity analysis formula is obtained to identify the reliability-dominated scenario in the source-load multi-period time series scenario.

2. The multi-dimensional weak link identification method according to claim 1, characterized in that: The distribution system reliability indicators include the system average power outage duration SAIDI, the expected power shortage EENS, and the system average power outage number SAIFI.

3. The multi-dimensional weak link identification method according to claim 1, characterized in that: The obtaining of power system parameters and determining the type of impact of each component failure on the load node further includes: Based on the power transfer mode and power restoration time after the load node is out of power due to component failure, the four types of component failure impacts on the load node are determined; Based on the definition of the four types of component failure impact types, the four types of component failure impact correlation matrix are defined.

4. The multi-dimensional weak link identification method according to claim 1, characterized in that: The method further includes: determining the partial derivative of the time-invariant component fault parameter under a single load level based on the obtained distribution system reliability index to obtain a sensitivity analysis formula, calculating the sensitivity of the time-invariant component fault parameter under a single load level, and identifying the reliability weak component under a single load level. Obtain sensitivity analysis topology of time-invariant component fault parameters under a single load level; Based on the obtained topology, the sensitivity of the system average power outage duration SAIDI, expected power shortage EENS, and average power outage number SAIFI to the failure rate of components in different locations is calculated. The sensitivity values ​​of the same parameter are normalized and compared with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis. Based on the obtained topology, the sensitivity of the system average outage duration SAIDI and the expected power shortage EENS to the fault repair time of components at different locations, the section switch operation time, and the tie switch operation time are calculated. The sensitivity values ​​of the same parameter are normalized and compared with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis. Among them, the equipment failure rate λ under a single load level i The sensitivity analysis formula is: Equipment failure repair time μ under a single load level i The sensitivity analysis formula is: Sectional switch operation time t under single load level sw The sensitivity analysis formula is: Tie switch operation time t under single load level op The sensitivity analysis formula is: Where λ i represents the equipment failure rate of the i-th component, n j Indicates the number of users of the jth load node, column (A i )、column(B i ) and column(C i ) represent the i-th column of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC, respectively, row(A i )、row(B i ) and row(C i ) represents the i-th row of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC respectively.

5. The multi-dimensional weak link identification method according to claim 1, characterized in that: The method of determining the time-invariant component fault parameters in the source-load multi-period time sequence scenario by taking partial derivatives of the time-invariant component fault parameters based on the obtained distribution system reliability index to obtain a sensitivity analysis formula, analytically calculating the sensitivity of the time-invariant component fault parameters in the source-load multi-period time sequence scenario, and identifying the reliability weak components in the source-load multi-period time sequence scenario also includes: Based on the calculated distribution system reliability index, the partial derivative of the component fault parameter is obtained to obtain the sensitivity formula of the reliability index affected by the time-invariant component fault parameter. Calculate the sensitivity of the system average outage duration SAIDI, expected power shortage EENS, and system average outage frequency SAIFI to the failure rate of components in different locations. Homogenize the different data, retain the actual values, and then compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis. Calculate the sensitivity of the system average outage duration SAIDI and expected power shortage EENS to the fault repair time of components at different locations, the operation time of sectionalizing switches, the operation time of tie switches, and the time of island formation. Homogenize the different data, retain the actual values, and then compare them with the corresponding sensitivity analysis formula to obtain the weak link identification results and cause analysis. Among them, the equipment failure rate λ in the source-load multi-period time sequence scenario i The sensitivity analysis formula is: Equipment failure repair time μ in the source-load multi-period time series scenario i The sensitivity analysis formula is: Sectional switch operation time t in the source-load multi-period time sequence scenario sw The sensitivity analysis formula is: Tie switch operation time t in the source-load multi-period timing scenario op The sensitivity analysis formula is: Island formation time t in the source-load multi-period time series scenario isld The sensitivity analysis formula is: Where λ i represents the equipment failure rate of the i-th component, n j Indicates the number of users of the jth load node, column (A i )、column(B i ) and column(C i ) represent the i-th column of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC, respectively, row(A i )、row(B i ) and row(C i ) represents the i-th row of the fault impact correlation matrix FEIMA, FEIMB, and FEIMC respectively; P k represents the probability of the kth scenario occurring, SAIDI(k) represents the average system outage time of the kth scenario, and IRM(k) represents the island recovery matrix of the kth scenario.

6. The multi-dimensional weak link identification method according to claim 1, characterized in that: Based on the obtained distribution system reliability index, the partial derivative of the time-invariant component fault parameters is obtained, including: Where β represents the fault parameter of a time-invariant component in the distribution system, E(F) represents the expected value of a reliability index of the distribution system, and f k represents the reliability index corresponding to the kth typical scenario, P k represents the probability of the kth typical scenario occurring, and K represents the number of typical running scenarios after reduction.

7. The multi-dimensional weak link identification method according to claim 1, characterized in that: The time-varying scenario parameters under the source-load multi-period time series scenario are defined in the time-varying scenario matrix. Time-varying scene matrix It is defined as, Where, Indicates the value of the parameter of the i-th scenario at the j-th node in the k-th scenario, N represents the number of nodes in the distribution system, L k,j , DG k,j and ESS k,j They represent the distributed power output and energy storage component output of the j-th node in the k-th scenario respectively.

8. The multi-dimensional weak link identification method according to claim 1, characterized in that: According to the basic principle of the finite difference method, the sensitivity of the time-varying scenario parameters in the source-load multi-period time series scenario is analyzed and calculated, and the sensitivity analysis formula is obtained. According to the basic principle of the finite difference method, the SAIDI(k)-TVS sensitivity calculation formula in the source-load multi-period time series scenario is: The calculation formula for EENS(k)-TVS sensitivity in the source-load multi-period time series scenario is: In the formula, the row vector Represents the average annual power outage times of N nodes, row vector Represents the annual average power outage time of N nodes, the row vector Represents the power shortage of N nodes, the row vector P 1×K =[P1,P2,…,P K ] represents the probability of K scenes appearing, μ(i) j LP ens(i) j LP They represent the power outage time and power shortage of the jth node in the i-th scenario respectively; SAIFI(k), SAIDI(k) and EENS(k) represent the average power outage times, average power outage time and expected power shortage of the system in the k-th scenario respectively, n represents the row vector composed of the number of users of N load nodes arranged in the order of number, Ncons represents the total number of users of the distribution system, [row k (L)] T represents the transposition of the row vector formed by all load demands in the kth scenario, ||ens(k) LP ||1 means taking a vector 1 norm of ; IRM(k) is the island recovery matrix.

9. A multi-dimensional weak link identification device for reliability improvement, characterized in that: include: The parameter acquisition module is used to determine the components and dominant scenarios of the weak links that need to be analyzed, and obtain component failure parameters and scenario parameters; The component fault parameters include time-invariant component fault parameters under a single load level and time-invariant component fault parameters under a source-load multi-period timing scenario; The scenario parameters are time-varying scenario parameters in a source-load multi-period time series scenario; A reliability index acquisition module is used to obtain power system parameters, determine the type of component failure impact of each component failure on the load node, and calculate the reliability index of the distribution system under the influence of the component failure; A module for identifying weak reliability components under a single load level is used to obtain the partial derivative of the time-invariant component fault parameters under a single load level based on the obtained distribution system reliability index, obtain a sensitivity analysis formula, calculate the sensitivity of the time-invariant component fault parameters under a single load level, and identify weak reliability components under a single load level. The time-invariant component fault parameters under a single load level include equipment failure rate, equipment failure repair time, section switch operation time, and tie switch operation time; The module for identifying weak reliability components in source-load multi-period time-series scenarios is used to determine the partial derivatives of the time-invariant component fault parameters based on the obtained distribution system reliability index, analytically calculate the sensitivity of the time-invariant component fault parameters in source-load multi-period time-series scenarios, and identify weak reliability components in source-load multi-period time-series scenarios. The reliability-dominated scenario identification module in the source-load multi-period time series scenario is used to define the time-varying scenario matrix for the time-varying scenario parameters in the source-load multi-period time series scenario. Based on the basic principles of the finite difference method, it analyzes and calculates the sensitivity of the time-varying scenario parameters in the source-load multi-period time series scenario, and performs reliability-dominated scenario identification in the source-load multi-period time series scenario.

10. The multi-dimensional weak link identification device according to claim 9, characterized in that: The time-invariant component fault parameters in the source-load multi-period time sequence scenario include equipment failure rate, equipment failure repair time, section switch operation time, tie switch operation time, and island formation time; The time-varying scenario parameters include load node demand, distributed power output, and energy storage element output.

Images