A comprehensive assessment method and system for the resilience of power distribution networks

Abstract

The present invention relates to a comprehensive evaluation method and system for the resilience of a distribution network, the method comprising: obtaining distribution network parameters, and conducting an evaluation according to a preset comprehensive evaluation system for the resilience of the distribution network; the comprehensive evaluation system for the resilience of the distribution network comprises primary evaluation indicators and secondary evaluation indicators, and each primary evaluation indicator is provided with a corresponding secondary evaluation indicator; the primary evaluation indicators include perception, resilience, defense, resilience, coordination and learning, and based on the evaluation results of each secondary evaluation indicator, a comprehensive calculation is conducted according to the weights of each secondary evaluation indicator and each primary evaluation indicator, to obtain a comprehensive evaluation result of the resilience of the distribution network. Compared with the prior art, the present invention targets six categories of key features of a resilient power grid, focuses on the three functions of distribution network situational awareness, disturbance response and self-improvement capabilities, and can establish a more comprehensive and refined comprehensive evaluation system under resilience requirements, and improve the accuracy and reliability of the evaluation results.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network resilience assessment, and in particular to a distribution network resilience comprehensive assessment method and system. Background Art

[0002] At present, the gradual depletion of traditional fossil energy and the increasingly severe global environmental situation have made it urgent to vigorously develop renewable energy including wind power and photovoltaics. There is an urgent need to build a stronger and more intelligent power system, especially to enhance its ability to cope with low-probability high-risk events. The concept of "resilient power grid" came into being in this context.

[0003] In view of the concept of resilient power grid, foreign scholars have conducted preliminary research on relevant theoretical methods and required key technologies, while domestic scholars have also innovatively proposed the connotation and concept of resilient power grid based on the development status and characteristics of my country's power system, including pointing out the six key characteristics of resilient power grid: "perception", "resilience", "defense", "resilience", "synergy" and "learning". Existing research on power grid resilience assessment mainly focuses on the characteristics of narrow resilience, that is, the resilience, defense and recovery of the system's ability to cope with sudden accidents before, during and after the accident. The document "Research on the resilience assessment method of island integrated energy system under earthquake disasters" (Li Xue, Sun Tingkai, Hou Kai, Jiang Tao, Chen Houhe, Li Guoqing, Jia Hongjie. Proceedings of the CSEE, 2020, 40(17): 5476-5493) proposed three indicators of robustness, rapidity and redundancy to evaluate the island integrated energy system; the document "Shaping and evaluation modeling of system safety resilience" (Huang Lang, Wu Chao, Wang Bing. China Occupational Safety Science and Technology, 2016, 12(12): 15-21) constructed a quantitative assessment model of safety resilience from three dimensions: system constituent elements, correlation and resilience function; the document "Evaluation and construction method of power grid resilience capacity for emergencies" (Xiao Zhiwen, Wang Guoqing, Zhu Jianming, et al. Systems Engineering - Theory and Practice, 2019, 39(10): 26372645) used performance curves to quantitatively evaluate the resilience of the power grid throughout the entire accident cycle. However, the evaluation research considering the three broad resilience characteristics of perception, coordination and learning is relatively limited. Currently, there are only some discussions on perception, but most of them focus on combining situational awareness technology to evaluate the specific operating performance of the power grid. The evaluation of the respective implementation effects of the broad characteristics under resilience requirements is still in the initial development stage.

[0004] As one of the main goals of the next stage of power system development, the resilient grid is huge and complex, and is highly coupled with other systems in the resilient city, making it more difficult to accurately assess the resilience of the grid. In order to quickly and accurately grasp the current resilience level of the distribution network and provide reliable data support for the formulation of subsequent operational decisions, it is of great research significance to analyze the distribution network resilience assessment method and establish a comprehensive evaluation system. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art, that is, the research on the assessment of the three broad resilience characteristics of perception, coordination and learning is relatively limited. Currently, there are only some discussions on perception, but most of them focus on combining situational awareness technology to assess the specific operating performance of the power grid. The evaluation of the respective implementation effects of the broad characteristics under resilience requirements is still in the initial development stage. A comprehensive assessment method and system for the resilience of a distribution network is provided.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A comprehensive evaluation method for the resilience of a distribution network comprises the following steps:

[0008] Obtain distribution network parameters and evaluate them according to a preset comprehensive evaluation system for distribution network resilience; the comprehensive evaluation system for distribution network resilience includes primary evaluation indicators and secondary evaluation indicators, and each of the primary evaluation indicators has a corresponding secondary evaluation indicator;

[0009] The first-level evaluation indicators of the comprehensive evaluation system for distribution network resilience include perception, resilience, defense, recovery, coordination and learning;

[0010] The secondary evaluation indicators corresponding to the perception capability include one or more of smart meter coverage, weak node observability, grid measurement redundancy, average transmission delay, situation visibility, and distribution automation system operation indicators;

[0011] The secondary evaluation indicators corresponding to the synergy include one or more of the distribution line connection rate, the proportion of coordinated flexible loads, the distribution network transfer rate and the local clean energy consumption rate;

[0012] The secondary evaluation indicators corresponding to the learning ability include one or more of the expected error between situation prediction data and actual data and the proportion of repairable vulnerabilities of the post-disaster perception system;

[0013] Based on the evaluation results of each secondary evaluation indicator, a comprehensive calculation is performed according to the preset weights of each secondary evaluation indicator and the weights of each primary evaluation indicator to obtain a comprehensive evaluation result of the resilience of the distribution network.

[0014] Furthermore, among the secondary evaluation indicators corresponding to the perception, the calculation expression of the smart meter coverage rate A1 is:

[0015]

[0016] Where n sm Represents the number of smart meters in the power grid area, n m Indicates the total number of electricity meters in the power grid area;

[0017] The calculation expression of the weak node observability A2 is:

[0018]

[0019] Where n wm represents the number of significant weak nodes, n w Indicates the total number of weak nodes;

[0020] The calculation expression of the power grid measurement redundancy A3 is:

[0021]

[0022] Where n pm Represents the number of observable nodes, n p Indicates the total number of grid nodes.

[0023] Furthermore, in the secondary evaluation index corresponding to the perception, the calculation expression of the average transmission delay A4 is:

[0024]

[0025] In the formula, t mi represents the time when the electricity quantity collected by meter i is measured, t ui Indicates the time when the measurement data of meter i is updated to the database, n m Indicates the total number of electricity meters in the power grid area;

[0026] The calculation expression of the situation visibility A5 is:

[0027]

[0028] Where n is the number of blocks divided into the situation map, N i is the number of nodes in block i, N i The arithmetic mean of

[0029] The calculation expression of the distribution automation system operation index A6 is:

[0030] A6=α aor ×P aor +αrs ×P rs +α rc ×P rc +α fac ×P fac

[0031] Where P aor represents the average online rate of distribution automation terminals, P rs Indicates the remote control success rate, P rc Indicates the accuracy rate of remote signaling action, P fac represents the success rate of feeder automation, α aor , α rs , α rc , α fac is the weight corresponding to each indicator, α aor +α rs +α rc +α fac =1.

[0032] Furthermore, among the secondary evaluation indicators corresponding to the synergy, the calculation expression of the distribution line connection rate E1 is:

[0033]

[0034] Where n l,H is the total number of 35-110 kV high-voltage lines in the region, n tl,H is the number of 35~110kV high-voltage interconnection lines in the region, n l,L is the total number of 10(20)kV low-voltage lines in the area, n tl,L is the number of 10(20)kV low-voltage interconnection lines in the region, α l,H , α l,L are the index weights of high-voltage lines and low-voltage lines respectively, α l,H +α l,L =1;

[0035] The calculation expression of the distribution network transfer rate E2 is:

[0036]

[0037] Where n l,tl is the number of transferable lines within the distribution network, n l is the total number of lines.

[0038] Furthermore, in the secondary evaluation index corresponding to the synergy force, the calculation expression of the coordinable flexible load ratio E3 is:

[0039]

[0040] In the formula, SFL is the adjustable flexible load peak, S L,max The maximum load of the annual grid supply;

[0041] The calculation expression of the local clean energy consumption rate E4 is:

[0042]

[0043] Where P oi is the net amount of electricity received from outside the region, P oa is the amount of electricity agreed outside the region, P co is the local clean energy grid-connected electricity, P cg Generates electricity from local clean energy.

[0044] Furthermore, in the secondary evaluation index corresponding to the learning ability, the calculation expression of the expected error F1 between the situation prediction data and the actual data is:

[0045]

[0046] Where N z,t is the total number of measurements predicted by the perception system at time t, is the estimated value of the i-th measurement obtained by the perception system at time t, x true,i is the true value of the system state corresponding to the i-th measurement at time t;

[0047] The calculation expression of the repairable vulnerability ratio F2 of the post-disaster perception system is:

[0048]

[0049] Where n vf is the total number of vulnerabilities found in the perception system after the disaster, n vr The number of repairable vulnerabilities in the perception system after a disaster.

[0050] Furthermore, the secondary evaluation indicators corresponding to the strain force include one or more of voltage transient rate, power flow over-limit rate, voltage harmonic distortion rate, frequency deviation rate, power distribution requirement rate, N-1 verification pass rate, active reserve rate and topology integrity.

[0051] Furthermore, the calculation expression of the voltage transient rate B1 is:

[0052]

[0053]

[0054] Where V i,max is the transient upper voltage limit of node i, V i,min is the transient lower limit voltage of node i, V i(t) refers to the voltage of node i at the current transient moment, and T is the statistical period;

[0055] The calculation expression of the power flow over-limit rate B2 is:

[0056]

[0057]

[0058] In the formula, S i,max is the rated upper limit of branch i, S i (t) is the power flow of branch i at time t, n l is the total number of power grid branches;

[0059] The calculation expression of the voltage harmonic distortion rate B3 is:

[0060]

[0061] Where V 1,i is the effective value of the fundamental voltage at node i, V k,i is the effective value of the kth harmonic voltage at node i;

[0062] The calculation expression of the frequency deviation rate B4 is:

[0063]

[0064] Where, f is the current frequency of the system, f N is the rated frequency, Δf th is the frequency deviation limit.

[0065] Furthermore, the calculation expression of the power distribution requirement rate B5 is:

[0066]

[0067] Where P a is the actual total power consumption of users in the resilient grid, P N Total rated frequency for resilient grid users;

[0068] The calculation expression of the N-1 verification pass rate B6 is:

[0069]

[0070] Where n t is the total number of substations within the power grid, n t(N-1) is the number of substations that have passed N-1 verification, n l is the total number of lines within the power grid, n l(N-1) is the number of lines that pass N-1 verification, α t , αl are the weights of substation and line indicators respectively, α t +α l =1;

[0071] The calculation expression of the active reserve rate B7 is:

[0072]

[0073] Where P r is the active reserve capacity of the power grid, P r.lim It is the active reserve capacity limit of the power grid;

[0074] The calculation expression of the topological integrity B8 is:

[0075]

[0076] In the formula, s i (t) is the operating status of line i at time t. When line i is operating normally, s i (t)=1, when line i is out of service, s i (t) = 0, T is the statistical period.

[0077] Furthermore, the secondary evaluation index corresponding to the defense capability includes one or more of a performance index, a power distribution load ratio and a power grid failure rate.

[0078] Furthermore, the calculation expression of the performance index C1 is:

[0079]

[0080] Where P o is the original grid capacity before the disaster, P d is the grid capacity after taking active defense measures, P low is the capacity when the grid performance drops to the lowest level, t low is the time when the grid performance drops to the lowest level, t d Time to take proactive defensive measures;

[0081] The calculation expression of the distribution network capacity-load ratio C2 is:

[0082]

[0083] In the formula, S T is the total capacity of distribution network substation equipment, S L,max The maximum load of the annual grid supply;

[0084] The calculation expression of the power grid failure rate C3 is:

[0085]

[0086] Where n pe is the total number of power equipment in the power grid, p fault,i is the failure probability of device i.

[0087] Furthermore, the secondary evaluation index corresponding to the resilience includes one or more of average power outage time of users, fault self-healing rate, black start success rate and average pre-scheduled power outage time of the system.

[0088] Furthermore, the calculation expression of the user's average power outage time D1 is:

[0089]

[0090] In the formula, t cut is the duration of the power outage caused by the ith fault, n ucut,i is the number of users who had power outages due to the i-th fault, n u is the total number of grid users;

[0091] The calculation expression of the fault self-healing rate D2 is:

[0092]

[0093] Where n ush,i is the number of users who self-recovered from the i-th failure, n ufault,i is the total number of users affected by the i-th failure;

[0094] The calculation expression of the black start success rate D3 is:

[0095]

[0096] Where n ubs,i The number of users whose power supply is restored through black start for the i-th fault, n ufault,i is the total number of users affected by the i-th failure.

[0097] Furthermore, the calculation expression of the average scheduled power outage time D4 of the system is:

[0098]

[0099] In the formula, t pc,i For the i-th scheduled power outage time, n upc,i is the number of users scheduled for power outage for the i-th time, n u is the total number of grid users.

[0100] Furthermore, the process of setting the weights of the secondary evaluation index and the primary evaluation index includes the following steps:

[0101] Subjective weighting step: Subjectively weight each indicator;

[0102] Objective weighting steps: objectively weight each indicator;

[0103] Comprehensive weight optimization step: Obtain the weight vectors obtained by each weighting method in the subjective weighting step and the objective weighting step; thereby calculating the overall set weight of each weighting method, and the calculation expression of the set weight is:

[0104]

[0105] Where M is the total number of weighting methods, u m =(u m1 ,u m2 ,…,u mN ) is the weight vector of the mth weighting method, u mn is the weight of the nth indicator obtained by the mth weighting method, N is the total number of indicators, d n is the set weight;

[0106] The relative entropy between the result of each weighting method and the set weight is calculated. The calculation expression of the relative entropy is:

[0107]

[0108] In the formula, h(u m ,d) is the relative entropy between the weight vector of the mth weighting method and the set weight;

[0109] According to the relative entropy, the preference coefficient of each weighting method is calculated, and the calculation expression of the preference coefficient is:

[0110]

[0111] In the formula, a m is the preference coefficient of the mth weighting method;

[0112] The comprehensive indicator weight coefficient of each indicator is calculated according to the preference coefficient. The calculation expression of the comprehensive indicator weight coefficient is:

[0113]

[0114] In the formula, w n is the comprehensive indicator weight coefficient of the nth indicator.

[0115] Furthermore, the subjective weighting step includes: using a binomial coefficient method to subjectively weight each indicator, and the binomial coefficient method includes the following steps:

[0116] M experts compare the N evaluation indicators in pairs and independently obtain the importance ranking of the indicator set O m, take the average of the rankings of each expert to get the average importance ranking of the nth indicator. The calculation expression of the average importance ranking of the nth indicator is:

[0117]

[0118] In the formula, O(x n ) is the average importance ranking of the nth indicator, O m (n) is the importance ranking of the mth indicator to the nth indicator;

[0119] Rearrange the N evaluation indicators in order from small to large according to the average importance, and obtain a new indicator sequence:

[0120]

[0121] In the formula, x1, x2, …, x N is the sorted evaluation index;

[0122] Make a symmetric ordering of the index set:

[0123] x N ,…,x2,x1,x3,…,x N-1

[0124] According to the symmetrical sorting indicator set, the indicators are numbered from left to right again, recorded as i, and the subjective weight of each indicator can be calculated. The calculation expression of the subjective weight of each indicator is:

[0125]

[0126] In the formula, u i is the subjective weight of the indicator numbered i, It is the calculation result of the indicator combination.

[0127] Furthermore, before the objective weighting step is performed, the step further includes: normalizing each indicator, and the normalization process is specifically as follows:

[0128] The normalized expression of benefit-based indicators is:

[0129]

[0130] In the formula, x mn is the actual calculated value of the nth indicator of the mth candidate solution, y mn ′ is the normalized value of the benefit index of the nth index of the mth candidate solution, x n,min is the minimum value of the nth indicator, x n,max is the maximum value of the nth indicator, and the candidate solution is the actual indicator value obtained by adopting the indicator;

[0131] The normalized expression of cost-based indicators is:

[0132]

[0133] In the formula, y mn ″ is the normalized value of the cost-type indicator of the nth indicator of the mth objective weighting scheme.

[0134] Furthermore, the objective weighting step includes: objectively weighting each indicator using an anti-entropy weight method, and the calculation process of the anti-entropy weight method includes:

[0135] Calculate the anti-entropy value h of each indicator n , the anti-entropy value h n The calculation expression is:

[0136]

[0137]

[0138] In the formula, y mn is the normalized value of the nth indicator of the mth objective weighting scheme, and M is the total number of options to be selected;

[0139] The weight of each indicator is further determined according to the anti-entropy value, and the calculation expression of the indicator weight is:

[0140]

[0141] In the formula, u n is the indicator weight of the nth indicator, and N is the total number of indicators.

[0142] Furthermore, the objective weighting step includes: objectively weighting each indicator using the CRITIC method, and the calculation process of the CRITIC method includes:

[0143] Calculate the redundant information entropy p of each indicator n , the redundant information entropy p n The calculation expression is:

[0144]

[0145]

[0146] In the formula, p n is the redundant information entropy of the nth indicator, y mn is the normalized value of the nth indicator of the mth objective weighting scheme, and M is the total number of options to be selected;

[0147] Use the normalized matrix to calculate the covariance between columns and the coefficient of variation s of the indexn , thereby calculating the correlation coefficient between the indicators, the calculation expression of the correlation coefficient between the indicators is:

[0148]

[0149]

[0150]

[0151] In the formula, is the normalized value of the nth indicator, s n is the coefficient of variation of the nth indicator, N is the total number of indicators, For the nth item and the nth * The correlation coefficient of the item, For nth * The index value of the item, is the index value y n and The covariance between For nth * The coefficient of variation of the item;

[0152] Evaluate the amount of information contained in each indicator. The calculation expression of the information is:

[0153]

[0154] In the formula, i n is the information content of the nth indicator;

[0155] The indicator weight is determined according to the amount of information, and the calculation expression of the indicator weight is:

[0156]

[0157] In the formula, u n is the indicator weight of the nth indicator.

[0158] The present invention also provides a comprehensive assessment system for the resilience of a distribution network, comprising a memory and a processor, wherein the memory stores a computer program, and the processor calls the computer program to execute the steps of the method described above.

[0159] Compared with the prior art, the present invention has the following advantages:

[0160] (1) Based on the characteristics of resilient power grids, the present invention constructs a comprehensive evaluation index system for the resilience of distribution networks including 27 micro-indicators from six macro-resilience measurement dimensions, namely, perception, strain, defense, recovery, synergy, and learning. In particular, the setting of secondary indicators of perception, synergy, and learning is innovatively set and proposed by the present invention, which can achieve a more comprehensive comprehensive evaluation of the distribution network's situational awareness, disturbance response, and self-improvement capabilities under resilience requirements. The present invention targets the six categories of key characteristics of resilient power grids, focusing on the three functions of distribution network situational awareness, disturbance response, and self-improvement capabilities, and can establish a more comprehensive and refined comprehensive evaluation system under resilience requirements, thereby improving the accuracy and reliability of the evaluation results.

[0161] (2) The present invention adopts a comprehensive weight optimization method based on relative entropy, combining the binomial coefficient method, anti-entropy weight method, and correlation weight method to comprehensively weight the indicator system, which can realize the weight calculation combining subjective and objective factors and improve the accuracy and reliability of the evaluation results.

[0162] (3) The present invention mainly considers the hardware configuration and operation level and the degree of grasp of the system status in the evaluation of perception. The specific indicators of perception have the following technical effects:

[0163] Smart meter coverage: As an important component of the advanced measurement system, smart meters are the basic terminal equipment that enables the situational awareness system to measure, store, and calculate power measurements and communicate with data centers in both directions. Their coverage reflects the basic awareness level of a resilient power grid.

[0164] Visibility of weak nodes: When a high-risk event occurs, there is a greater probability that the power performance of weak nodes will decline. The degree of grasp of the real-time status of weak nodes can reflect the ability of the resilient grid perception system to respond to potential disasters, eliminate hidden dangers in a timely manner and prevent further expansion of faults;

[0165] Grid measurement redundancy: Different from the observability of weak nodes, which mainly reflects the response capability of the grid sensing system under disaster conditions, grid measurement redundancy reflects whether the sensing system of the resilient grid can fully grasp the grid operation status and assist other systems in making decisions under normal conditions. It can serve as a supplement to the observability of weak nodes.

[0166] Average transmission delay: The average transmission delay is defined as the average time required for the resilient power grid operation data to be transmitted in the perception system and updated to the database, which reflects the real-time requirements of the resilient power grid perception system;

[0167] Situation visibility: Situation visibility reflects the visualization level of the perception system. Convenient and clear situation information is conducive to the operation and management personnel to quickly analyze power events and make corresponding decisions when responding to disasters and accidents. The visualization level of the situation map is evaluated by the node distribution uniformity. The smaller the node distribution uniformity, the more uniform the sample data distribution and the higher the visualization level.

[0168] Distribution automation system operation indicators: As a collection of power grid data collection and monitoring systems, geographic information systems, demand-side management systems and other systems, the distribution automation system integrates functions such as power data monitoring and analysis and remote control. Unlike several previous system configuration indicators, the distribution automation system operation indicators mainly reflect the perceived results of the decision-making execution accuracy and execution efficiency of the resilient power grid, and are real-time dynamic indicators.

[0169] (4) The synergy index set by the present invention represents the ability of the resilient power grid to reasonably and efficiently utilize internal and external resources and jointly concentrate its strength to resist disturbance events. The synergy-related index can be used to improve and reflect the distribution network's ability to control internal and external defense resources, multi-dimensional system information, etc.; the specific synergy indexes have the following technical effects:

[0170] Distribution line connection rate: As the most intuitive and basic connection method, the direct connection of distribution lines provides flexibility for the dispatching and coordination of physical systems within and outside the region. At the same time, the increase of coupling elements also provides higher measurement redundancy for the distribution network.

[0171] Coordinated flexible load ratio: As an important component of the active distribution network, flexible load is responsible for maintaining power balance and disaster coordinated recovery; the coordinated flexible load ratio determines the peak capacity of the resilient grid's coordinated anti-interference ability, reflecting the synergistic effect of flexible load on grid management

[0172] Distribution network transfer rate: Line transfer can effectively reduce the amount of load loss in the case of local faults in the resilient power grid, give full play to the system's coordinated disaster response capabilities to a greater extent, and also provide greater flexibility for distribution network risk situation assessment and decision-making effect simulation;

[0173] Local clean energy consumption rate: The amount of clean energy consumed can comprehensively reflect the synergistic effect of a resilient grid. Unlike the proportion of coordinated flexible loads, it dynamically characterizes the level of coordinated balance of electric power within and outside the region and the flexible anti-interference capability.

[0174] (5) The learning ability set in the present invention serves as a common support for other features, characterizing the ability of the resilient power grid to self-correct and improve from historical experience and to combine new technologies to improve innovation. It can be used to reflect the ability of the resilient distribution network to autonomously correct errors and continuously improve. The specific indicators of the learning ability have the following technical effects:

[0175] Expected error between situation prediction data and actual data: By combining historical operation data and post-accident risk situation review analysis, the error between the predicted data and actual data of the resilient power grid can be obtained. The expected error and change trend can effectively reflect the self-correction and autonomous improvement capabilities of the distribution network.

[0176] Proportion of vulnerabilities that can be repaired by the post-disaster perception system: The proportion of vulnerabilities that can be repaired by the post-disaster perception system directly reflects the vulnerability identification, error correction and self-learning capabilities of the resilient power grid system. BRIEF DESCRIPTION OF THE DRAWINGS

[0177] Figure 1 A schematic diagram of a flow chart of a binomial coefficient method provided in an embodiment of the present invention;

[0178] Figure 2 A schematic diagram of a flow chart of an anti-entropy weight method provided in an embodiment of the present invention;

[0179] Figure 3 A schematic diagram of a CRITIC method provided in an embodiment of the present invention;

[0180] Figure 4 A schematic diagram of a comprehensive weight optimization process based on relative entropy provided in an embodiment of the present invention;

[0181] Figure 5 A schematic diagram of a comprehensive resilience assessment process for a distribution network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0182] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0183] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0184] Example 1

[0185] This embodiment provides a method for comprehensively evaluating the resilience of a distribution network, including the following steps:

[0186] Obtain distribution network parameters and conduct an evaluation based on a preset comprehensive evaluation system for distribution network resilience; the comprehensive evaluation system for distribution network resilience includes primary evaluation indicators and secondary evaluation indicators, and each primary evaluation indicator has a corresponding secondary evaluation indicator;

[0187] The first-level evaluation indicators of the comprehensive evaluation system for distribution network resilience include perception, resilience, defense, recovery, coordination and learning;

[0188] The secondary evaluation indicators corresponding to perception include one or more of smart meter coverage, weak node observability, grid measurement redundancy, average transmission delay, situation visibility and distribution automation system operation indicators; among them, smart meter coverage, weak node observability and grid measurement redundancy are used to reflect the perception level of the distribution network edge side; average transmission delay, situation visibility and distribution automation system operation indicators are used to reflect the perception level of the distribution network control center;

[0189] The secondary evaluation indicators corresponding to the strain force include one or more of voltage transient rate, power flow over-limit rate, voltage harmonic distortion rate, frequency deviation rate, power distribution demand rate, N-1 verification pass rate, active reserve rate and topology integrity; among them, voltage transient rate, power flow over-limit rate, voltage harmonic distortion rate and frequency deviation rate are used to reflect the electrical quantity parameter indicators of the distribution network operation process; power distribution demand rate, N-1 verification pass rate, active reserve rate and topology integrity are used to reflect the macro planning indicators of the distribution network;

[0190] The secondary evaluation indicators corresponding to the defense capability include one or more of the performance index, the distribution load ratio, and the grid failure rate;

[0191] The secondary evaluation indicators corresponding to resilience include one or more of the average power outage time of users, fault self-healing rate, black start success rate and average system pre-scheduled power outage time; among which, the average system pre-scheduled power outage time is used to reflect the pre-accident warning and maintenance level of the distribution network; the average power outage time of users, fault self-healing rate and black start success rate are used to reflect the post-accident recovery level of the distribution network;

[0192] The secondary evaluation indicators corresponding to synergy include one or more of the distribution line connection rate, the proportion of coordinated flexible loads, the distribution network transfer rate and the local clean energy consumption rate; among them, the distribution line connection rate and the distribution network transfer rate are used to reflect the physical line coordination level of the distribution network; the proportion of coordinated flexible loads and the local clean energy consumption rate are used to reflect the flexible resource coordination level of the distribution network.

[0193] The secondary evaluation indicators corresponding to learning ability include one or more of the expected error between situation prediction data and actual data and the proportion of repairable vulnerabilities in the post-disaster perception system;

[0194] Based on the evaluation results of each secondary evaluation indicator, a comprehensive calculation is performed according to the preset weights of each secondary evaluation indicator and the weights of each primary evaluation indicator to obtain a comprehensive evaluation result of the resilience of the distribution network.

[0195] There is no limit to the number of secondary evaluation indicators corresponding to the above-mentioned perception, adaptability, defense, resilience, coordination and learning ability, and they can be arbitrarily selected from the above-mentioned secondary evaluation indicators. It is preferred to adopt all the corresponding secondary evaluation indicators provided in this embodiment. An optimal implementation method is described in detail below.

[0196] 1. Construction of a comprehensive evaluation system for distribution network resilience

[0197] Based on the relationship between the key characteristics of resilient power grids and the principles for selecting micro-evaluation indicators, the actual resilience level of the distribution network is comprehensively evaluated from six dimensions: perception, resilience, defense, recovery, coordination, and learning. A comprehensive evaluation index system for distribution network resilience is established as shown in Table 1.

[0198] Table 1 Comprehensive evaluation index system for distribution network resilience

[0199]

[0200]

[0201] The following is a detailed introduction to each secondary evaluation indicator:

[0202] 1. Perception

[0203] Based on the principles of convenience and practicality of data acquisition, the evaluation of perception mainly considers its hardware configuration and operation level and the degree of grasp of system status. Indicators related to the perception level of resilient power grids, such as algorithm efficiency and risk situation prediction, which are difficult to directly quantify in evaluation, will be indirectly reflected by subsequent indicators of other dimensions.

[0204] 1) Smart meter coverage

[0205] As an important part of the advanced measurement system, smart meters are basic terminal equipment that enables the situational awareness system to measure, store, and calculate power measurements and communicate with data centers in both directions. Their coverage rate reflects the basic perception level of the resilient power grid. The calculation formula is as follows:

[0206]

[0207] Where n sm Represents the number of smart meters in the power grid area, n m Indicates the total number of electricity meters in the grid area.

[0208] 2) Visibility of weak nodes

[0209] When a high-risk event occurs, there is a greater probability that the power performance of weak nodes will decline. The degree of grasp of the real-time status of weak nodes can reflect the ability of the resilient grid perception system to respond to potential disasters, promptly eliminate hidden dangers and prevent further expansion of faults. The node weakness is determined based on the load-system short-circuit capacity ratio. The calculation formula for the load-system short-circuit capacity ratio is as follows:

[0210]

[0211] In the formula, E Pi , Z Pi is the Thevenin equivalent circuit parameter of the power grid at the node, U Li , Z Li is the Thevenin equivalent circuit parameter of the load at the node. The larger the load-system short-circuit capacity ratio is, the smaller the voltage margin that the node can withstand is. When this index is higher than the preset threshold, the node is considered a weak node. When the voltage phasor of the node can be directly measured or calculated, the node is observable. The calculation formula for the observability of a weak node is as follows:

[0212]

[0213] Where n wm represents the number of significant weak nodes, n w Indicates the total number of weak nodes.

[0214] 3) Grid measurement redundancy

[0215] Different from the observability of weak nodes, which mainly reflects the response capability of the power grid sensing system under disaster conditions, the power grid measurement redundancy reflects whether the sensing system can fully grasp the power grid operation status and assist other systems in making decisions under normal conditions of the resilient power grid, which can serve as a supplement to the observability of weak nodes. The calculation formula of power grid measurement redundancy is as follows:

[0216]

[0217] Where n pm Represents the number of observable nodes, n p Indicates the total number of grid nodes.

[0218] 4) Average transmission delay

[0219] The average transmission delay is defined as the average time required for the resilient power grid operation data to be transmitted in the perception system and updated to the database, reflecting the real-time requirements of the resilient power grid perception system. The calculation formula is as follows:

[0220]

[0221] In the formula, t mirepresents the time when the electricity quantity collected by meter i is measured, t ui Indicates the time when the measurement data of meter i is updated to the database.

[0222] 5) Situational Visibility

[0223] Situation visibility reflects the visualization level of the perception system. Convenient and clear situation information is conducive to the operation and management personnel to quickly analyze power events and make corresponding decisions when responding to disasters and accidents. The visualization level of the situation map is evaluated by the node distribution uniformity. The smaller the node distribution uniformity, the more uniform the sample data distribution and the higher the visualization level. The specific calculation formula is as follows:

[0224]

[0225] Where n is the number of blocks divided into the situation map, N i is the number of nodes in block i, N i The arithmetic mean of .

[0226] 6) Distribution automation system operation indicators

[0227] As a collection of power grid data collection and monitoring systems, geographic information systems, demand-side management systems, and other systems, the distribution automation system integrates functions such as power data monitoring and analysis and remote control. Different from the previous system configuration indicators, the distribution automation system operation indicators mainly reflect the perception of the decision-making execution accuracy and execution efficiency of the resilient power grid. They are real-time dynamic indicators and include four sub-items: average online rate of distribution automation terminals, remote control success rate, remote signaling action accuracy, and feeder automation success rate.

[0228] The calculation expression of the distribution automation system operation index A6 is:

[0229] A6=α aor ×P aor +α rs ×P rs +α rc ×P rc +α fac ×P fac

[0230] Where P aor represents the average online rate of distribution automation terminals, P rs Indicates the remote control success rate, P rc Indicates the accuracy rate of remote signaling action, P fac represents the success rate of feeder automation, α aor , α rs , α rc , α fac is the weight corresponding to each indicator, αaor +α rs +α rc +α fac =1;

[0231] Preferably, the specific calculation formula for the optimal weight configuration of the distribution automation system operation index A6 given in this embodiment is as follows:

[0232] A6=0.25×P aor +0.25×P rs +0.2×P rc +0.3×P fac (7)

[0233] Where P aor represents the average online rate of distribution automation terminals, P rs Indicates the remote control success rate, P rc Indicates the accuracy rate of remote signaling action, P fac Indicates the success rate of feeder automation.

[0234] The coefficients in the distribution automation system operation index A6 can be adjusted according to actual conditions without specific limitation. This embodiment provides the optimal coefficients of the distribution automation system operation index A6.

[0235] 2. Resilience

[0236] Resilience reflects the ability of a resilient power grid to resist disturbances, maintain its own performance, and formulate risk plans before an extreme event occurs. Resilience-related indicators can be used to reflect the reliability and efficiency of the power grid in daily operation decisions, as well as the effectiveness of risk prediction and pre-accident deployment.

[0237] 1) Voltage transient rate

[0238] Voltage transient rate is mainly used to reflect the frequency and amplitude of voltage sag (or surge) during normal operation of the power grid. A lower voltage transient rate reflects the system's stronger transient stability and anti-interference ability, and also reflects the power grid's monitoring and early warning capabilities for transient disturbances. The voltage transient rate calculation formula is as follows:

[0239]

[0240]

[0241] Where V i,max is the transient upper voltage limit of node i, V i,min is the transient lower limit voltage of node i, V i (t) refers to the voltage of node i at the current transient moment, and T is the statistical period.

[0242] 2) Current flow rate

[0243] Frequent power flow over-limit will reduce the anti-interference ability of the power grid, which is not conducive to risk control before disasters. Maintaining a low frequency of power flow over-limit reflects the real-time power flow monitoring and risk situation prediction capabilities of the resilient power grid. The formula for calculating the power flow over-limit rate is as follows:

[0244]

[0245]

[0246] In the formula, S i,max is the rated upper limit of branch i, S i (t) is the power flow of branch i at time t, n l is the total number of power grid branches.

[0247] 3) Voltage harmonic distortion rate

[0248] The harmonic distortion of voltage reduces the power supply quality within the power grid. Excessive distortion may even cause heating failures in transformers and power lines. A lower voltage harmonic distortion rate meets the requirements of users in resilient cities for high power quality, is also conducive to suppressing small disturbances before accidents, and reflects the power quality monitoring capability of the distribution network. The voltage harmonic distortion rate calculation formula is as follows:

[0249]

[0250] Where V 1,i is the effective value of the fundamental voltage at node i, V k,i is the effective value of the kth harmonic voltage at node i.

[0251] 4) Frequency deviation rate

[0252] The frequency deviation rate is a steady-state power indicator of the power grid, which reflects the active safety performance of the system. Technologies such as flow monitoring and load forecasting and dispatching are conducive to reducing the frequency deviation of the resilient power grid. The calculation formula of the frequency deviation rate is as follows:

[0253]

[0254] Where, f is the current frequency of the system, f N is the rated frequency, Δf th The frequency deviation limit is 0.2Hz.

[0255] 5) Power distribution demand rate

[0256] The distribution demand rate is defined as the ratio of the actual operating power of users in a resilient power grid to the rated power. This value reflects the rationality and economy of the power resource allocation in the resilient power grid, and also reflects the grid's ability in flexible load scheduling and resource allocation. The distribution demand rate calculation formula is as follows:

[0257]

[0258] Where P a is the actual total power consumption of users in the resilient grid, P N Total rated frequency for resilient grid users.

[0259] 6) N-1 verification pass rate

[0260] The N-1 pass rate includes the N-1 pass rate of substations and the N-1 pass rate of transmission lines within the power grid. As the most intuitive indicator to characterize the reliability and risk warning capabilities of the power grid, it also reflects the ability of the resilient power grid to perceive risk situations and simulate anticipated accidents.

[0261] The calculation expression of N-1 check pass rate B6 is:

[0262]

[0263] Where n t is the total number of substations within the power grid, n t(N-1) is the number of substations that have passed N-1 verification, n l is the total number of lines within the power grid, n l(N-1) is the number of lines that pass N-1 verification, α t , α l are the weights of substation and line indicators respectively, α t +α l =1;

[0264] Preferably, the calculation formula for the optimal weight configuration of the N-1 verification pass rate given in this embodiment is as follows:

[0265]

[0266] Where n t is the total number of substations within the power grid, n t(N-1) is the number of substations that have passed N-1 verification, n l is the total number of lines within the power grid, n l(N-1) is the number of lines that pass the N-1 check.

[0267] 7) Active power reserve rate

[0268] Different from the distribution demand rate, the active reserve rate focuses more on the pre-arrangement of emergency resources. An appropriate amount of active reserve is conducive to maintaining system performance in risk accidents, reducing losses caused by disasters, and meeting the economic efficiency of normal system operation. The specific calculation formula is as follows:

[0269]

[0270] Where Pr is the active reserve capacity of the power grid, P r.lim It is the active reserve capacity limit of the power grid, which is generally taken as 0.1 times the maximum system load.

[0271] 8) Topological integrity

[0272] The integrity of the grid topology greatly affects its vulnerability to disasters. Reducing the number of outages and downtime is conducive to building a stronger, more reliable and resilient grid. The calculation formula for topological integrity is as follows:

[0273]

[0274] In the formula, s i (t) is the operating status of line i at time t. When line i is operating normally, s i (t)=1, when line i is out of service, s i (t)=0.

[0275] 3. Defense

[0276] Defense focuses on the ability of a resilient power grid to proactively resist damage from disasters and reduce the overall impact of an event by using internal and external defense resources during an extreme event. Defense-related indicators can be used to reflect the operational efficiency of the power grid under extreme conditions and the response rate under variable conditions.

[0277] 1) Performance index

[0278] In resilience assessment, available performance curves can be used to characterize the dynamic changes of resilient power grids during extreme events, and grid capacity is selected as the performance indicator.

[0279] The performance index is defined as the product of active defense speed and active defense effect. This index also reflects the real-time response capability of perception during disasters, the ability to predict the development of disaster situations, and the ability to evaluate defense resources. The specific calculation formula is as follows:

[0280]

[0281] Where P o is the original grid capacity before the disaster, P d is the grid capacity after taking active defense measures, P low is the capacity when the grid performance drops to the lowest level, t low is the time when the grid performance drops to the lowest level, t d Time to take proactive defensive measures.

[0282] 2) Distribution network capacity ratio

[0283] The distribution network load ratio is defined as the ratio of the total capacity of substation equipment in the distribution network area to the annual maximum load. A certain transformer spare capacity is conducive to improving the adaptability of the power grid during disasters and bringing greater flexibility to the dispatch of defense resources; at the same time, an excessively large load ratio not only increases the investment cost and construction period of the resilient power grid, but also reduces the economic efficiency of power grid operation. How to reduce the distribution network load ratio while meeting reliability depends on more accurate load forecasting and operation control technology of the distribution network. The distribution network load ratio calculation formula is as follows:

[0284]

[0285] In the formula, S T is the total capacity of distribution network substation equipment, S L,max It is the maximum annual grid load.

[0286] 3) Grid failure rate

[0287] In order to minimize the degradation of system performance and the scope of disturbance caused by disaster events, it is necessary to prevent the occurrence of large-scale cascading failures. Using monitoring and early warning functions to reduce the failure probability of each component of the power grid can improve the resilience of the resilient power grid in high-risk events. The formula for calculating the power grid failure rate considering the role of each component is as follows:

[0288]

[0289] Where n pe is the total number of power equipment in the power grid, p fault,i is the failure probability of equipment i, which can be obtained by combining historical data with situation prediction technology.

[0290] 4. Resilience

[0291] Resilience considers the emergency protection and gradual restoration of power grid performance after an accident, and timeliness is the most important consideration for resilience. Resilience-related indicators can be used to reflect the power grid's capabilities in fault location, maintenance and repair, remote control, and decision-making simulation evaluation.

[0292] 1) Average power outage time for users

[0293] Power outage time is the most direct indicator of the resilience level of a resilient power grid after a disturbance. Technologies such as fault location, fault data recording, and disturbance factor analysis can effectively reduce the average power outage time. The formula for calculating the average power outage time for users is as follows:

[0294]

[0295] In the formula, t cut is the duration of the power outage caused by the ith fault, n ucut,i is the number of users who had power outages due to the i-th fault, n uis the total number of grid users.

[0296] 2) Fault self-healing rate

[0297] Fault self-healing is mainly reflected in the self-repair ability of the power grid when facing minor disturbances or the early stage of extreme events. Its cycle includes three levels: monitoring and early warning, diagnosis and analysis, and automatic repair. A high-level resilient power grid can warn and diagnose its inducing factors when the fault is still in the "potential" stage, and use remote control to achieve fault self-healing attempts. The formula for calculating the fault self-healing rate is as follows:

[0298]

[0299] Where n ush,i is the number of users who self-recovered from the i-th failure, n ufault,i is the total number of users affected by the i-th failure.

[0300] 3) Black start success rate

[0301] As a key measure for emergency protection after an extreme event, the formulation of the optimal plan for black start depends largely on the distribution network's estimation of the system state under the fault set and the operation effect of the plan in a simulated environment such as a twin system. The black start success rate can effectively reflect the initial recovery capability of the resilient power grid after a disaster. The calculation formula is as follows:

[0302]

[0303] Where n ubs,i The number of users whose power supply is restored through black start for the i-th fault.

[0304] 4) Average scheduled power outage time of the system

[0305] Pre-arranged power outage means that the power department has a plan to shut down the relevant area for maintenance or capacity expansion, which is suitable for delayed fault repair that does not cause damage in a short time. Pre-arranged power outage is mainly determined based on indicators such as failure rate and fault rate. The formula for calculating the average pre-arranged power outage time of the system is as follows:

[0306]

[0307] In the formula, t pc,i For the i-th scheduled power outage time, n upc,i The number of users scheduled for power outage for the i-th time.

[0308] 5. Synergy

[0309] Synergy characterizes the ability of a resilient power grid to rationally and efficiently utilize internal and external resources and to jointly concentrate its strength to resist disturbance events. Like perception, it is a basic feature that serves the three core features. Indicators related to synergy can be used to improve and reflect the distribution network's ability to control internal and external defense resources, multi-dimensional system information, etc.

[0310] 1) Distribution line connection rate

[0311] As the most intuitive and basic communication method, direct connection of distribution lines provides flexibility for the scheduling and coordination of physical systems within and outside the region. At the same time, the increase in coupling elements also provides higher measurement redundancy for the distribution network.

[0312] The calculation expression of the distribution line connection rate E1 is:

[0313]

[0314] Where n l,H is the total number of 35-110 kV high-voltage lines in the region, n tl,H is the number of 35~110kV high-voltage interconnection lines in the region, n l,L is the total number of 10(20)kV low-voltage lines in the area, n tl,L is the number of 10(20)kV low-voltage interconnection lines in the region, α l,H , α l,L are the index weights of high-voltage lines and low-voltage lines respectively, α l,H +α l,L =1;

[0315] Preferably, the calculation formula for the optimal weight configuration of the distribution network connection rate given in this embodiment is as follows:

[0316]

[0317] Where n l,H is the total number of 35-110 kV high-voltage lines in the region, n tl,H is the number of 35~110kV high-voltage interconnection lines in the region, n l,L is the total number of 10(20)kV low-voltage lines in the area, n tl,L It is the number of 10(20)kV low voltage interconnection lines in the area.

[0318] The coefficients in the calculation formula of the distribution network connection rate can be adjusted according to actual conditions and are not specifically limited. This embodiment provides the optimal coefficients of the calculation formula of the distribution network connection rate.

[0319] 2) Distribution network transfer rate

[0320] Line transfer can effectively reduce the number of load losses in a resilient power grid in the event of a local fault, giving full play to the system's ability to coordinate disaster response to a greater extent, while also providing greater flexibility for distribution network risk assessment and decision-making effect simulation. The distribution network transfer rate calculation formula is as follows:

[0321]

[0322] 3) Adjustable flexible load ratio

[0323] As an important component of the active distribution network, flexible loads are responsible for maintaining power balance and disaster coordinated recovery. The proportion of flexible loads that can be coordinated determines the peak capacity of the resilient grid's coordinated anti-interference ability, reflecting the synergistic effect of flexible loads on grid management. The calculation formula is as follows:

[0324]

[0325] In the formula, S FL It is a flexible load peak that can be coordinated.

[0326] Where n l,tl It refers to the number of transferable supply lines within the distribution network.

[0327] 4) Local clean energy consumption rate

[0328] The amount of clean energy consumed can comprehensively reflect the synergistic effect of the resilient grid. Different from the proportion of coordinated flexible loads, it dynamically characterizes the coordinated balance level of power and electricity within and outside the region and the flexible anti-interference ability. The specific calculation formula is as follows:

[0329]

[0330] Where P oi is the net amount of electricity received from outside the region, P oa is the amount of electricity agreed outside the region, P co is the local clean energy grid-connected electricity, P cg Generates electricity from local clean energy.

[0331] 6. Learning ability

[0332] Learning ability, as a common support for other characteristics, represents the ability of a resilient power grid to self-correct and improve from historical experience and to combine new technologies to improve innovation. Indicators related to learning ability can be used to reflect the ability of a resilient distribution network to self-correct and continuously improve.

[0333] 1) Expected error between situation prediction data and actual data

[0334] Combining historical operation data and post-accident risk situation review analysis, the error between the predicted data and the actual data of the resilient power grid can be obtained. The error expectation and change trend can effectively reflect the self-correction and autonomous improvement capabilities of the distribution network. The specific calculation formula is as follows:

[0335]

[0336] Where N z,t is the total number of measurements predicted by the perception system at time t, is the estimated value of the i-th measurement obtained by the perception system at time t, x true,i is the true value of the system state corresponding to the i-th measurement at time t.

[0337] 2) Proportion of repairable vulnerabilities in post-disaster perception systems

[0338] The proportion of vulnerabilities that can be repaired by the post-disaster perception system directly reflects the vulnerability identification, error correction and self-learning capabilities of the resilient power grid system. The calculation formula is as follows:

[0339]

[0340] Where n vf is the total number of vulnerabilities found in the perception system after the disaster, n vr The number of repairable vulnerabilities in the perception system after a disaster.

[0341] 2. Weighting method for comprehensive assessment of power grid resilience

[0342] The weighting methods can be divided into subjective weighting methods and objective weighting methods according to their basis. The use of any one of the methods alone may lead to one-sided evaluation. Therefore, the present invention uses a comprehensive weighting optimization method based on relative entropy to optimize the subjective and objective weights obtained by the binomial coefficient method, the anti-entropy weight method and the CRITIC method to realize the weight calculation of the comprehensive evaluation of power grid resilience.

[0343] 1. Subjective weighting method

[0344] like Figure 1 As shown, the subjective weighting method of this embodiment adopts the binomial coefficient method. Compared with the Delphi method and AHP, the binomial coefficient method is simpler to calculate and comprehensively considers the opinions of all parties to avoid ignoring minority opinions. The binomial coefficient method flow chart is as follows Figure 1 As shown, the weighting steps are as follows:

[0345] S101: M experts compare the N evaluation indicators in pairs and independently obtain the importance ranking of the indicator set O m The smaller the ranking value, the more important the indicator. The average ranking of each expert is taken to get the average importance ranking of the nth indicator. The formula is as follows:

[0346]

[0347] S102: Rearrange the N evaluation indicators in ascending order according to their average importance to obtain a new indicator sequence:

[0348]

[0349] S103: Make a symmetrical sort of the indicator set, and calculate the binomial coefficient based on the sorting result:

[0350] x N ,…,x2,x1,x3,…,x N-1 (33)

[0351] S104: Number each indicator from left to right again according to the symmetrically sorted indicator set, recorded as i. The binomial coefficient can be determined according to the numbering result to calculate the subjective weight of each indicator, and the formula is as follows:

[0352]

[0353] In the formula, u i is the subjective weight of the indicator numbered i, It is the calculation result of the indicator combination.

[0354] 2. Objective empowerment method

[0355] The subjective weighting method relies on the subjective cognition of the evaluation experts on the evaluation index set, and there is the possibility of random deviation and one-sidedness. It must be used in combination with the objective weighting method that relies on objective data. The objective weighting methods selected in this invention are the anti-entropy weight method and the CRITIC method. The anti-entropy weight method is based on the concept of information entropy, which can effectively evaluate the quality of information provided by the indicators for decision-making, and avoids the extreme situation of too small weighting in the entropy weight method. At the same time, considering that the key features in the resilient power grid are highly coupled, the CRITIC method suitable for weighting highly correlated indicators is used to improve the effectiveness of objective weighting.

[0356] Assume that there are M options, each of which includes N evaluation indicators, and x mn It represents the actual calculated value of the nth indicator of the mth scheme. Since the actual calculated value ranges of various micro-indicators vary greatly and have different dimensions, it is not conducive to determining their objective weights. Therefore, each indicator is first normalized. The specific formula is as follows:

[0357] 1) Benefit-type indicators, that is, the indicator value is positively correlated with the ability being evaluated.

[0358]

[0359] 1) Cost-based indicators, that is, the indicator value is negatively correlated with the ability being evaluated.

[0360]

[0361] The normalized index values ​​y mn They are dimensionless constants ranging from 0 to 1, and the larger the index value, the stronger the ability of the scheme. The specific weight calculation process of the anti-entropy weight method and the CRITIC method will be introduced below.

[0362] 3. Anti-entropy weight method

[0363] The flowchart of the anti-entropy weight method is as follows: Figure 2 As shown, the specific steps are as follows:

[0364] S201: Obtain normalized index;

[0365] S202: Calculate the anti-entropy value h of each indicator n , the formula is as follows:

[0366]

[0367]

[0368] S203: Further determine the weight of each indicator according to the anti-entropy value, the formula is as follows:

[0369]

[0370] 4. CRITIC method

[0371] The empowerment process of CRITIC Act is as follows Figure 3 As shown, the specific steps are as follows:

[0372] S301: Obtain normalized index;

[0373] S302: Calculate the redundant information entropy p of each indicator n , the formula is as follows:

[0374]

[0375]

[0376] S303: Using the covariance between columns of the normalized matrix and the coefficient of variation s of the index n Calculate the correlation coefficient between the indicators, the formula is as follows:

[0377]

[0378]

[0379]

[0380] In the formula, is the normalized value of the nth indicator, s n is the coefficient of variation of the nth indicator, N is the total number of indicators, For the nth item and the nth * The correlation coefficient of the term, y n* For nth * The index value of the item, is the index value y n and The covariance between For nth * The coefficient of variation of the item.

[0381] S304: Evaluate the amount of information contained in each indicator. The formula is as follows:

[0382]

[0383] S305: Determine the indicator weight according to the amount of information:

[0384]

[0385] 5. Comprehensive weight optimization

[0386] After obtaining the subjective and objective weighting results, consider the determination method of subjective and objective preference coefficients based on relative entropy. The comprehensive weight optimization process based on relative entropy is as follows: Figure 4 As shown, the optimization steps are as follows:

[0387] S401: each weight vector u obtained by M weighting methods m =(u m1 ,u m2 ,…,u mN ), based on the principle of relative entropy, the set weight is obtained according to each weight vector through the optimization model. The set weight is an intermediate variable in the process of obtaining the comprehensive weight.

[0388] The specific optimization model can refer to the prior art, such as the following model formula:

[0389]

[0390] This embodiment directly gives the set weight d of the nth secondary evaluation index n The calculation formula is as follows:

[0391]

[0392] S402: Calculate the relative entropy between the result of each weighting method and the set weight to characterize the closeness between the result of the weighting method and the set weight. The formula is as follows:

[0393]

[0394] S403: Obtain the preference coefficient of the weighting method based on the closeness. The greater the closeness, the greater the role played by the method in determining the comprehensive weight, and the higher the preference coefficient. The calculation formula is as follows:

[0395]

[0396] S404: Obtain the comprehensive indicator weight coefficient combining subjective and objective weighting according to the preference coefficient:

[0397]

[0398] Therefore, according to the contents of the above sections, the comprehensive evaluation process of distribution network resilience is obtained as follows: Figure 5 As shown, the specific steps include:

[0399] S1: Construct a comprehensive evaluation index system for distribution network resilience;

[0400] S2: Calculate the index values ​​of various micro-indicators in the comprehensive evaluation index system of distribution network resilience based on the parameter data of the distribution network;

[0401] S3: Calculate the subjective weight and objective weight of each first-level evaluation index and second-level evaluation index;

[0402] S4: Optimize the subjective weight and objective weight of each indicator through comprehensive weight to obtain the final weight;

[0403] S5: Obtain the final evaluation result according to the final weight of the index and the calculated index value.

[0404] This embodiment also provides a distribution network resilience comprehensive assessment system, including a memory and a processor, wherein the memory stores a computer program, and the processor calls the computer program to execute the steps of the distribution network resilience comprehensive assessment method as described above.

[0405] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A comprehensive evaluation method for the resilience of a distribution network, characterized in that: The following steps are involved: Obtain distribution network parameters and evaluate them according to a preset comprehensive evaluation system for distribution network resilience; the comprehensive evaluation system for distribution network resilience includes primary evaluation indicators and secondary evaluation indicators, and each of the primary evaluation indicators has a corresponding secondary evaluation indicator; The first-level evaluation indicators of the comprehensive evaluation system for distribution network resilience include perception, resilience, defense, recovery, coordination and learning; The secondary evaluation indicators corresponding to the perception capability include one or more of smart meter coverage, weak node observability, grid measurement redundancy, average transmission delay, situation visibility, and distribution automation system operation indicators; The secondary evaluation indicators corresponding to the synergy include one or more of the distribution line connection rate, the proportion of coordinated flexible loads, the distribution network transfer rate and the local clean energy consumption rate; The secondary evaluation indicators corresponding to the learning ability include one or more of the expected error between situation prediction data and actual data and the proportion of repairable vulnerabilities of the post-disaster perception system; Based on the evaluation results of each secondary evaluation indicator, a comprehensive calculation is performed according to the preset weights of each secondary evaluation indicator and the weights of each primary evaluation indicator to obtain a comprehensive evaluation result of the resilience of the distribution network; Among the secondary evaluation indicators corresponding to the perception, the calculation expression of the smart meter coverage rate A1 is: Where n sm Represents the number of smart meters in the power grid area, n m Indicates the total number of electricity meters in the power grid area; The calculation expression of the weak node observability A2 is: Where n wm represents the number of significant weak nodes, n w Indicates the total number of weak nodes; The calculation expression of the power grid measurement redundancy A3 is: Where n pm Represents the number of observable nodes, n p Indicates the total number of power grid nodes; Among the secondary evaluation indicators corresponding to the perception, the calculation expression of the average transmission delay A4 is: Where, t mi Indicates the time when the electricity quantity collected by meter i is measured, t ui Indicates the time when the measurement data of meter i is updated to the database, n m Indicates the total number of electricity meters in the power grid area; The calculation expression of the situation visibility A5 is: Where n is the number of blocks divided into the situation map, N i is the number of nodes in block i, N i The arithmetic mean of The calculation expression of the distribution automation system operation index A6 is: A6=a aor ×P aor +a rs ×P rs +a rc ×P rc +a fac ×P fac Where P aor represents the average online rate of distribution automation terminals, P rs Indicates the remote control success rate, P rc Indicates the accuracy rate of remote signaling action, P fac represents the success rate of feeder automation, α aor , α rs , α rc , α fac is the weight corresponding to each indicator, α aor +α rs +α rc +α fac =1; Among the secondary evaluation indicators corresponding to the synergy, the calculation expression of the distribution line connection rate E1 is: Where n l,H is the total number of 35-110 kV high-voltage lines in the region, n tl,H is the number of 35~110kV high-voltage interconnection lines in the region, n l,L is the total number of 10(20)kV low-voltage lines in the area, n tl,L is the number of 10(20)kV low-voltage interconnection lines in the region, α l,H , α l,L are the index weights of high-voltage lines and low-voltage lines respectively, α l,H +α l,L =1; The calculation expression of the distribution network transfer rate E2 is: Where n l,tl is the number of transferable lines within the distribution network, n l is the total number of lines; Among the secondary evaluation indicators corresponding to the synergy force, the calculation expression of the coordinable flexible load ratio E3 is: In the formula, S FL is the adjustable flexible load peak, S L,max The maximum load of the annual grid supply; The calculation expression of the local clean energy consumption rate E4 is: Where P oi is the net amount of electricity received from outside the region, P oa is the amount of electricity agreed outside the region, P co is the local clean energy grid-connected electricity, P cg generating electricity for local clean energy; Among the secondary evaluation indicators corresponding to the learning ability, the calculation expression of the expected error F1 between the situation prediction data and the actual data is: Where N z,t is the total number of measurements predicted by the perception system at time t, is the estimated value of the i-th measurement obtained by the perception system at time t, x true,i is the true value of the system state corresponding to the i-th measurement at time t; The calculation expression of the repairable vulnerability ratio F2 of the post-disaster perception system is: Where n vf is the total number of vulnerabilities found in the perception system after the disaster, n vr The number of repairable vulnerabilities in the perception system after a disaster.

2. A comprehensive evaluation method for the resilience of a distribution network according to claim 1, characterized in that: The secondary evaluation indicators corresponding to the strain force include one or more of voltage transient rate, power flow over-limit rate, voltage harmonic distortion rate, frequency deviation rate, power distribution requirement rate, N-1 verification pass rate, active power reserve rate and topology integrity.

3. A comprehensive evaluation method for the resilience of a distribution network according to claim 2, characterized in that: The calculation expression of the voltage transient rate B1 is: Where V i,max is the transient upper voltage limit of node i, V i,min is the transient lower limit voltage of node i, V i (t) refers to the voltage of node i at the current transient moment, and T is the statistical period; The calculation expression of the power flow over-limit rate B2 is: In the formula, S i,max is the rated upper limit of branch i, S i (t) is the power flow of branch i at time t, n l is the total number of power grid branches; The calculation expression of the voltage harmonic distortion rate B3 is: Where V 1,i is the effective value of the fundamental voltage at node i, V k,i is the effective value of the kth harmonic voltage at node i; The calculation expression of the frequency deviation rate B4 is: Where, f is the current frequency of the system, f N is the rated frequency, Δf th is the frequency deviation limit.

4. A comprehensive evaluation method for the resilience of a distribution network according to claim 2, characterized in that: The calculation expression of the power distribution requirement rate B5 is: Where P a is the actual total power consumption of users in the resilient grid, P N Total rated frequency for resilient grid users; The calculation expression of the N-1 verification pass rate B6 is: Where n t is the total number of substations within the power grid, n t(N-1) is the number of substations that have passed N-1 verification, n l is the total number of lines within the power grid, n l(N-1) is the number of lines that pass N-1 verification, α t , α l are the weights of substation and line indicators respectively, α t +α l =1; The calculation expression of the active reserve rate B7 is: Where P r is the active reserve capacity of the power grid, P r.lim It is the active reserve capacity limit of the power grid; The calculation expression of the topological integrity B8 is: In the formula, s i (t) is the operating status of line i at time t. When line i is operating normally, s i (t)=1, when line i is out of service, s i (t) = 0, T is the statistical period.

5. A comprehensive evaluation method for the resilience of a distribution network according to claim 1, characterized in that: The secondary evaluation index corresponding to the defense capability includes one or more of a performance index, a power distribution load ratio, and a power grid failure rate.

6. A comprehensive evaluation method for the resilience of a distribution network according to claim 5, characterized in that: The calculation expression of the performance index C1 is: Where P o is the original grid capacity before the disaster, P d is the grid capacity after taking active defense measures, P low is the capacity when the grid performance drops to the lowest level, t low is the time when the grid performance drops to the lowest level, t d Time to take proactive defensive measures; The calculation expression of the distribution capacity load ratio C2 is: In the formula, S T is the total capacity of distribution network substation equipment, S L,max The maximum load of the annual grid supply; The calculation expression of the power grid failure rate C3 is: Where n pe is the total number of power equipment in the power grid, p fault,i is the failure probability of device i.

7. A comprehensive evaluation method for the resilience of a distribution network according to claim 1, characterized in that: The secondary evaluation indicators corresponding to the resilience include one or more of the average power outage time of users, the fault self-healing rate, the black start success rate and the average pre-scheduled power outage time of the system.

8. A comprehensive evaluation method for the resilience of a distribution network according to claim 7, characterized in that: The calculation expression of the user's average power outage time D1 is: Where, t cut,i is the duration of the power outage caused by the ith fault, n ucut,i is the number of users who experienced power outages due to the i-th fault, n u is the total number of grid users; The calculation expression of the fault self-healing rate D2 is: Where n ush,i is the number of users who self-recovered from the i-th failure, n ufault,i is the total number of users affected by the i-th failure; The calculation expression of the black start success rate D3 is: Where n ubs,i The number of users whose power supply is restored through black start for the i-th fault, n ufault,i is the total number of users affected by the i-th failure.

9. A comprehensive evaluation method for the resilience of a distribution network according to claim 7, characterized in that: The calculation expression of the average scheduled power outage time D4 of the system is: Where, t pc,i For the i-th scheduled power outage time, n upc,i is the number of users scheduled for power outage for the i-th time, n u is the total number of grid users.

10. A comprehensive evaluation method for the resilience of a distribution network according to claim 1, characterized in that: The process of setting the weights of the secondary evaluation index and the primary evaluation index includes the following steps: Subjective weighting step: Subjectively weight each indicator; Objective weighting steps: objectively weight each indicator; Comprehensive weight optimization step: Obtain the weight vectors obtained by each weighting method in the subjective weighting step and the objective weighting step; thereby calculating the overall set weight of each weighting method, and the calculation expression of the set weight is: Where M is the total number of weighting methods, u m =(u m1 ,u m2 ,…,u mN ) is the weight vector of the mth weighting method, u mn is the weight of the nth indicator obtained by the mth weighting method, N is the total number of indicators, d n is the set weight; The relative entropy between the result of each weighting method and the set weight is calculated. The calculation expression of the relative entropy is: In the formula, h(u m ,d) is the relative entropy between the weight vector of the mth weighting method and the set weight; According to the relative entropy, the preference coefficient of each weighting method is calculated, and the calculation expression of the preference coefficient is: In the formula, a m is the preference coefficient of the mth weighting method; The comprehensive indicator weight coefficient of each indicator is calculated according to the preference coefficient. The calculation expression of the comprehensive indicator weight coefficient is: In the formula, w n is the comprehensive indicator weight coefficient of the nth indicator.

11. A comprehensive evaluation method for the resilience of a distribution network according to claim 10, characterized in that: The subjective weighting step includes: using the binomial coefficient method to subjectively weight each indicator, and the binomial coefficient method includes the following steps: M experts compare the N evaluation indicators in pairs and independently obtain the importance ranking of the indicator set O m , take the average of the rankings of each expert to get the average importance ranking of the nth indicator. The calculation expression of the average importance ranking of the nth indicator is: In the formula, O(x n ) is the average importance ranking of the nth indicator, O m (n) is the importance ranking of the mth indicator to the nth indicator; Rearrange the N evaluation indicators in order from small to large according to the average importance, and obtain a new indicator sequence: In the formula, x1, x2, …, x N is the sorted evaluation index; Make a symmetric ordering of the index set: x N ,…,x2,x1,x3,…,x N-1 According to the symmetrical sorting indicator set, the indicators are numbered from left to right again, recorded as i, and the subjective weight of each indicator can be calculated. The calculation expression of the subjective weight of each indicator is: In the formula, u i is the subjective weight of the indicator numbered i, It is the calculation result of the indicator combination.

12. A comprehensive evaluation method for the resilience of a distribution network according to claim 10, characterized in that: Before the objective weighting step is performed, the method further includes: normalizing each indicator. The normalization process is specifically as follows: The normalized expression of benefit-based indicators is: In the formula, x mn is the actual calculated value of the nth indicator of the mth candidate solution, y mn ′ is the normalized value of the benefit index of the nth index of the mth candidate solution, x n,min is the minimum value of the nth indicator, x n,max is the maximum value of the nth indicator, and the candidate solution is the actual indicator value obtained by adopting the indicator; The normalized expression of cost-based indicators is: In the formula, y mn ″ is the normalized value of the cost-type indicator of the nth indicator of the mth objective weighting scheme.

13. A comprehensive evaluation method for the resilience of a distribution network according to claim 12, characterized in that: The objective weighting step includes: objectively weighting each indicator using the anti-entropy weight method, and the calculation process of the anti-entropy weight method includes: Calculate the anti-entropy value h of each indicator n , the anti-entropy value h n The calculation expression is: In the formula, y mn is the normalized value of the nth indicator of the mth objective weighting scheme, and M is the total number of weighting methods; The weight of each indicator is further determined according to the anti-entropy value, and the calculation expression of the indicator weight is: In the formula, u n is the indicator weight of the nth indicator, and N is the total number of indicators.

14. A comprehensive evaluation method for the resilience of a distribution network according to claim 12, characterized in that: The objective weighting step includes: objectively weighting each indicator using the CRITIC method, and the calculation process of the CRITIC method includes: Calculate the redundant information entropy p of each indicator n , the redundant information entropy p n The calculation expression is: In the formula, p n is the redundant information entropy of the nth indicator, y mn is the normalized value of the nth indicator of the mth objective weighting scheme, and M is the total number of weighting methods; Use the normalized matrix to calculate the covariance between columns and the coefficient of variation s of the index n , thereby calculating the correlation coefficient between the indicators, the calculation expression of the correlation coefficient between the indicators is: In the formula, is the normalized value of the nth indicator, s n is the coefficient of variation of the nth indicator, N is the total number of indicators, For the nth item and the nth * The correlation coefficient of the item, For nth * The index value of the item, is the index value y n and The covariance between For nth * The coefficient of variation of the item; Evaluate the amount of information contained in each indicator. The calculation expression of this information is: In the formula, i n is the information content of the nth indicator; The indicator weight is determined according to the amount of information, and the calculation expression of the indicator weight is: In the formula, u n is the indicator weight of the nth indicator.

15. A comprehensive evaluation system for the resilience of a distribution network, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor calls the computer program to execute the steps of any one of the methods according to claims 1 to 14.

Images