Method and system for evaluating failure probability and vulnerability of power grid based on improved Pearl network

Abstract

The invention relates to the technical field of urban power grid safety, and provides a power grid failure probability and vulnerability assessment method and system based on an improved Pearl network. The method comprises the following steps: firstly, improving an input relationship, an output relationship, an event identifier M and a transition activation and execution rule in a traditional Pearl network to obtain an improved Pearl network of an urban power grid system; then, determining a transition failure probability of event transition in the urban power grid system based on the cloud model, determining a place failure probability of the urban power grid system through Monte Carlo simulation based on an improved Pearl network of the urban power grid system, and representing power grid accident evolution of the urban power grid system through the place failure probability; the method comprises the following steps: firstly, determining an overall power grid failure probability, secondly, determining the risk of a target place in an urban power grid system based on a cloud model, and finally, performing vulnerability assessment on the urban power grid system according to the power grid failure probability and the risk of the target place.

Description

Technical Field

[0001] The present application relates to the field of urban power grid security technology, and in particular to a power grid failure probability and vulnerability assessment method and system based on an improved Petri network. Background Art

[0002] As a complex power system, the stable operation of the urban power grid is directly related to the city's energy supply and the daily life of residents. It is particularly important to conduct effective failure probability analysis and vulnerability assessment of the urban power grid system, which can effectively prevent and reduce the impact of power grid failures and provide guarantees for the stable and reliable operation of the urban power grid.

[0003] At present, urban power grid failure probability analysis and vulnerability assessment mostly rely on traditional linear models and statistical methods. However, these methods have great limitations when facing the complexity, dynamics and nonlinearity of urban power grid systems. Especially when simulating concurrent events, time dependencies and complex system interactions, it is often difficult to capture the full picture of the urban power grid system, thus affecting the accuracy and reliability of the assessment results.

[0004] Therefore, there is an urgent need to provide a technical solution to the above-mentioned deficiencies in the prior art. Summary of the invention

[0005] The purpose of this application is to provide a method and system for evaluating power grid failure probability and vulnerability based on an improved Petri network, so as to solve or alleviate the problems existing in the above-mentioned prior art.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] The present application provides a method for evaluating the failure probability and vulnerability of a power grid based on an improved Petri network, which is used to evaluate the failure probability and vulnerability of a city power grid system, comprising: step S101, improving the input relationship, output relationship, event identifier M, transition activation and execution rules in a traditional Petri network to obtain an improved Petri network of the city power grid system;

[0008] Step S102: determining a power grid failure probability of the urban power grid system based on an improved Petri network of the urban power grid system;

[0009] Step S103: determining the danger of a target place in the urban power grid system based on the cloud model; wherein the target place is a place corresponding to the final state of the evolution of the event transition in the urban power grid system;

[0010] Step S104: performing a vulnerability assessment on the urban power grid system according to the power grid failure probability and the danger level of the target storage location.

[0011] Preferably, in step S101, the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri network are improved, including: the input library P of the urban power grid system i The input correlation matrix of the transition T and the output library P of the transition T to the urban power grid system j The output association matrix is ​​changed; wherein, the input association matrix is ​​allowed to have at most one arc from the input library P i To the transition T; the output association matrix allows at most one arc from the transition T to the output library P j ; The input association matrix represents the input library P i The output correlation matrix represents the input relationship from the transition T to the output library P. j The output relationship of; i, j are both positive integers;

[0012] Determine that the event identifier M of each place P of the urban power grid system is: P→{0, 1}, indicating the event state of the place P, wherein 0 indicates that the event corresponding to the place P has not occurred, and 1 indicates that the event corresponding to the place P has occurred;

[0013] Determine that the preset transition activation threshold is met in the urban power grid system, and the input library P i The corresponding event has occurred, the transition T is activated, and the input library P i The event state evolves to the output library P j The event status.

[0014] In step S101, according to the historical fault data and operation data of the urban power grid system, key factors affecting the evolution of failure events of the urban power grid system are extracted based on the Delphi method to construct an improved Petri network of the urban power grid system.

[0015] Preferably, step S102 comprises: determining the transition failure probability of event transition in the urban power grid system based on the cloud model; wherein the event transition represents the evolution of power grid accidents in the urban power grid system;

[0016] Based on the improved Petri net of the urban power grid system, the failure probability of the storage place of the urban power grid system is determined by Monte Carlo simulation;

[0017] Based on the transition failure probability and the library failure probability, a power grid failure probability of the urban power grid system is determined.

[0018] Preferably, the cloud model-based determination of the transition failure probability of the event transition in the urban power grid system comprises: determining a scale cloud of the failure language evaluation of the urban power grid system based on a forward cloud generator; wherein the failure language evaluation characterizes the failure probability of the event transition;

[0019] According to the failure language evaluation of the event transition defined by the ruler cloud, based on the reverse cloud generator, a judgment cloud of the event transition in the urban power grid system is generated to determine the transition activation threshold of the event transition, and whether the event transition in the urban power grid system is valid is judged based on the transition activation threshold.

[0020] Preferably, the determination of the scale cloud of the failure language evaluation of the urban power grid system based on the forward cloud generator includes: determining the interval values ​​of multiple cloud models of the scale cloud of the failure language evaluation of the urban power grid system based on the cloud model "3En" rule, and calculating the scale cloud diagram of the failure probability of the event transition through the forward cloud generator algorithm. Preferably, the determination of the failure probability of the place of the urban power grid system through Monte Carlo simulation based on the improved Petri network of the urban power grid system includes: based on Monte Carlo simulation, according to the determined transition activation condition function, cyclically running the improved Petri network until the number of running calculations of the improved Petri network is equal to the set number of Monte Carlo simulations Sim, and determining the ratio of the number of times the place event of the urban power grid system occurs to the set number of Monte Carlo simulations Sim as the place failure probability; wherein, the place event represents the state change of the urban power grid system.

[0021] Preferably, in step S103, the danger of the target place is determined according to a preset qualitative value language description and a danger scale cloud diagram of the final state of evolution.

[0022] Preferably, in step S104, the product of the power grid failure probability and the danger of the target storage place is calculated to determine the vulnerability of the urban power grid system;

[0023] Assigning values ​​to the failure probability of the power grid and the failure language level of the dangerousness of the target library respectively to construct a vulnerability assessment matrix of the urban power grid system; wherein the failure language level represents the probability of the evolution of the urban power grid accident;

[0024] Based on the vulnerability assessment matrix and according to the vulnerability of the urban power grid system, a vulnerability level of the urban power grid system is determined.

[0025] The embodiment of the present application also provides a power grid failure probability and vulnerability assessment system based on an improved Petri network, which is used to assess the failure probability and vulnerability of a city power grid system, including: a Petri network improvement unit, configured to improve the input relationship, output relationship, event identifier M, transition activation and execution rules in a traditional Petri network to obtain an improved Petri network of the city power grid system;

[0026] a failure judgment unit configured to determine a power grid failure probability of the urban power grid system based on an improved Petri network of the urban power grid system;

[0027] Danger judgment is configured to determine the danger of a target library in the urban power grid system based on a cloud model; wherein the target library is a library corresponding to the final state of the evolution of the event transition in the urban power grid system;

[0028] The vulnerability unit is configured to perform a vulnerability assessment on the urban power grid system according to the power grid failure probability and the danger of the target storage location.

[0029] Beneficial effects:

[0030] The method for evaluating the failure probability and vulnerability of a power grid based on an improved Petri net provided in the embodiment of the present application is used to evaluate the failure probability and vulnerability of an urban power grid system. First, the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri net are improved to obtain an improved Petri net of the urban power grid system. Then, based on the improved Petri net of the urban power grid system, the probability of power grid failure of the urban power grid system is determined. Next, based on the cloud model, the danger of the target location in the urban power grid system is determined. Finally, based on the power grid failure probability and the danger of the target location, the vulnerability of the urban power grid system is evaluated. Thus, a comprehensive and accurate urban power grid transition failure probability and vulnerability evaluation system is provided, which accurately identifies the vulnerable points of the power grid and proposes effective mitigation and improvement measures to improve the stability and reliability of the power grid system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. Among them:

[0032] Figure 1 A schematic flow chart of a method for evaluating power grid failure probability and vulnerability based on an improved Petri network according to some embodiments of the present application;

[0033] Figure 2 A schematic diagram of an improved execution rule of Petri network transition provided according to an embodiment of the present application;

[0034] Figure 3 A schematic diagram of a flow chart of failure probability analysis of an output library based on Monte Carlo simulation according to some embodiments of the present application;

[0035] Figure 4 A schematic diagram of the evolution of a secondary chain event scenario of a power grid transmission line tripping provided according to an embodiment of the present application;

[0036] Figure 5 A schematic diagram of the evolution of secondary chain events of power grid line tripping based on an improved Petri network according to an embodiment of the present application;

[0037] Figure 6 A schematic diagram of a scale cloud diagram of a failure probability of a transition provided according to an embodiment of the present application;

[0038] Figure 7 A schematic diagram of a cloud diagram of the value range of the transition T8 condition function F being true according to an embodiment of the present application;

[0039] Figure 8 A schematic diagram of a cloud diagram of a hazard scale of an evolutionary final state provided according to an embodiment of the present application;

[0040] Fig. 9 A schematic diagram of the structure of a power grid failure probability and vulnerability assessment system based on an improved Petri grid is provided according to some embodiments of the present application. DETAILED DESCRIPTION

[0041] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. Each example is provided by way of explanation of the present application and does not limit the present application. In fact, it will be clear to those skilled in the art that modifications and variations may be made in the present application without departing from the scope or spirit of the present application. For example, a feature shown or described as a part of an embodiment may be used in another embodiment to produce yet another embodiment. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the embodiments of the present invention should belong to the scope of protection of the embodiments of the present invention.

[0042] The method for evaluating the failure probability and vulnerability of a power grid based on an improved Petri network provided in the embodiment of the present application analyzes the failure probability of a city power grid system and evaluates its vulnerability. Figure 1 , Figure 2 As shown in FIG. 1 , the power grid failure probability and vulnerability assessment method based on the improved Petri network includes:

[0043] Step S101, improving the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri network to obtain an improved Petri network of the urban power grid system.

[0044] In this application, based on the historical fault data and operation data of the urban power grid, an improved Petri network for simulating the evolution of power grid events is constructed. First, the historical fault data and operation data of the urban power grid are collected and analyzed to extract the key factors affecting the evolution of failure events of the urban power grid system.

[0045] Among them, the historical fault data and operation data of the urban power grid are collected and analyzed, and the key factors affecting the evolution of failure events of the urban power grid system are extracted based on the Delphi method. Here, the scenario of secondary chain events of power grid transmission line tripping is taken as an example to illustrate. First, focus on collecting data related to transmission line tripping and its secondary chain events, including tripping event records, reclosing success and failure, power grid stability and grid structure, etc.; according to the specific tripping event case, based on the Delphi method, identify the interaction and influence of different factors, and then identify and construct the key event chain occurring in the tripping event. For example, tripping (S1) may lead to reclosing failure (S3), which in turn causes power grid grid structure damage (C5) or local stability problems (C6~C7). 10 ), analyze the conditions that trigger these events, such as the impact of delay protection (R1), inspection judgment (R2) and load reduction measures (R3, R6) on the event chain; convert the collected information into nodes and edges in the improved Petri net according to the improved Petri net rules.

[0046] Then, according to the specific characteristics of the urban power grid system and the uniqueness of the events in the urban power grid system, an improved Petri network suitable for the urban power grid system is formulated to accurately simulate the evolution of the power grid over time. Specifically, the input relationship and output relationship in the traditional Petri network are improved as follows: i The input correlation matrix of the transition T and the output library P of the transition T to the urban power grid system j The output association matrix of is changed; in which, at most one arc is allowed to be from the input library P in the input association matrix i To transition T; the output incidence matrix allows at most one arc from transition T to output place P j ; Input correlation matrix represents input library P i The input relationship to the transition T, the output correlation matrix represents the transition T to the output library P j The output relationship of ; i and j are both positive integers.

[0047] The event identifier M in the traditional Petri network is improved as follows: the event identifier M of each place P of the urban power grid system is determined as: P→{0, 1}, which represents the event state of place P. Among them, M: P→{0, 1} is a (1×n) vector, representing the number of tokens in each place. Here, the place has only 0 or 1 token, because the token represents that the event has occurred. Therefore, 0 tokens in the place represent that the event corresponding to place P has not occurred, and 1 token in the place represents that the event corresponding to place P has occurred. For the evolution process of the urban power grid accident, the initial representation M0 is (1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0), which means that at the beginning of the evolution, only the urban power grid accident represented by the first place has occurred.

[0048] The improvement of the transition activation and execution rules in the traditional Petri network is as follows: determine the transition activation threshold that meets the preset transition activation threshold in the urban power grid system and the input library P i The corresponding event has occurred, transition T is activated, and the input library P i The event state evolves to the output library P j The event status.

[0049] That is to say, transition T can only be activated if the threshold range of the transition is met and there is a token in its input place Pi. That is, for the chain evolution process, the evolution path can only be formed after the event represented by the previous input place occurs and the external conditions for the transition T to be successfully activated are met. i The event state evolves to the output library P j The event status.

[0050] When transition T is successfully activated, the value of the new event marker M′ is: M′(P i )=M(P i ), M′(P j )=1. Among them, P i is the input place of transition T, P j is the output place of transition T, that is, after transition T is successfully activated, its input place P i The token is transferred to the output place P j But the input library P i The token value does not disappear, because the existence of the token here represents whether the library event occurs or not. i The event represented by the library has occurred, so the number of tokens does not decrease, P j The event represented by the library has also occurred, so the token number changes from 0 to 1, such as Figure 3 shown.

[0051] In a specific example, the construction of the improved Petri network is explained by taking the scenario of secondary chain events of power transmission line tripping as an example. First, data related to transmission line tripping and its secondary chain events are collected, including: tripping event records, reclosing success and failure, power grid stability and grid structure, etc. By analyzing specific tripping event cases, the interaction and influence between different factors are determined, and the key time chain occurring in the tripping event is identified and constructed. For example, tripping (S1) may lead to reclosing failure (S3), which in turn causes power grid structure damage (C5) or local stability problems (C6-C7). 10 )(like Figure 4 As shown), the meaning of each node is shown in Table 1, which is as follows:

[0052] Table 1 Explanation of node numbers in scenario evolution diagram

[0053]

[0054]

[0055] Analyze the triggering events (tripping (S1), reclosing failure (S3), grid structure damage (C5), local stability problems (C6-C 10 )) conditions, such as the impact of delay protection (R1), inspection judgment (R2) and load reduction measures (R3; R6) on the event chain. The collected information is converted into nodes and edges in the improved Petri network according to the rules of the improved Petri network. For example, S1-S 17 , R1-R6, E1-E6, C1-C 11 Transformed into the improved Petri net library (P1-P 16 ) and transition (T1-T 15 ),like Figure 5 As shown in Table 2.

[0056] Table 2 Description of the evolution nodes of secondary chain events caused by power transmission line tripping

[0057]

[0058]

[0059] Step S102: determining the power grid failure probability of the urban power grid system based on the improved Petri network of the urban power grid system.

[0060] In this application, the failure probability of the power grid system is analyzed by using the causal relationship of the constructed Petri network, combined with the cloud model and Monte Carlo simulation, and the value range of the transition and the threshold for activating the transition are calculated by the cloud model to reflect the possibility and conditions of the transition failure. Then, according to the execution rules of the transition activation in the Petri network, multiple experiments are carried out through Monte Carlo simulation to obtain the statistical results of the evolution network of the failure event of the urban power grid system, so as to accurately calculate the failure probability of the power grid system under different evolution paths.

[0061] Specifically, based on the cloud model, the transition failure probability of event transition in the urban power grid system is determined; based on the improved Petri net of the urban power grid system, the failure probability of the library of the urban power grid system is determined through Monte Carlo simulation; and then, based on the transition failure probability and the failure probability of the library, the grid failure probability of the urban power grid system is determined. Among them, the event transition represents the evolution of the grid accident of the urban power grid system, that is, the dynamic change of the system state of the urban power grid system.

[0062] First, based on the cloud forward generator, the scale cloud for the language evaluation of urban power grid system failure is determined. Specifically, the scale cloud is used as the threshold of transition T in the evolution of urban power grid accident scenarios, and the probability of urban power grid transition failure represented by transition T is described in qualitative value language (i.e., failure language evaluation), and is judged into five language degree levels, namely: {extremely unlikely failure, less likely failure, half likely failure, more likely failure, extremely likely failure}.

[0063] In order to determine the failure probability of transition T (i.e., transition failure probability), five clouds are determined on the number field [0, 1] according to the golden section method, and the middle cloud is set as C0 (E x0 , E n0 , H e0 ), the adjacent clouds on the left and right are C -1 (E x-1 , E n-1 , H e-1 ), C +1 (E x+1 , E n+1 , H e+1 ), C -2 (E x-2 , E n-2 , H e-2 ), C +2 (E x+2 , E n+2 , H e+2 ); the algorithm formula for the digital features of each cloud is as follows:

[0064]

[0065] According to the “3En” rule of the cloud model, more than 99.73% of cloud droplets will fall on [E x -3E n , E x +3E n ], the interval values ​​of the five cloud models of the scale cloud for the language evaluation of urban power grid system failure are determined. The scale cloud diagram of the transition failure probability of transition T is calculated through the forward cloud generator algorithm.

[0066] Based on the system failure language evaluation (i.e., the transition failure probability analysis evaluation language value) level of transition T in the evolution of urban power grid accident scenarios defined by the ruler cloud and the interval values ​​corresponding to the five cloud models, the maximum and minimum values ​​of the failure probability of each evaluation unit are determined. Through the action of the reverse cloud generator, the quantitative analysis value is converted into a digital feature (E) that reflects the qualitative concept of the transition failure probability in the evaluation unit. x , E n , H e ), that is, the maximum value of the transition failure probability cloud model C max and minimum cloud model C min Then, according to the comprehensive cloud model, the final judgment cloud of the transition failure probability in each evaluation unit is established by calculation as the T1-T 15 The value range of the transition. Among them, the comprehensive cloud model is:

[0067]

[0068] In the comprehensive cloud model, the calculation between different cloud models under the same number field [0, 1] is obtained by the operation rules of the cloud model. Finally, through the forward cloud generator algorithm, the judgment cloud map based on the failure probability cloud map of the scale cloud transition is obtained.

[0069] Then, according to the failure language evaluation of event transition defined by the ruler cloud, the judgment cloud of event transition in the urban power grid system is generated based on the reverse cloud generator to determine the transition activation threshold of event transition. Taking the evolution of the risk scenario of secondary chain events caused by the tripping of power transmission lines in the power grid as an example, the ruler cloud and judgment cloud models are used to realize the occurrence possibility analysis of all transitions in the evolution diagram of the risk scenario of secondary chain events caused by the tripping of power transmission lines in the power grid. The value range of the transition is obtained by the judgment cloud, and the ruler cloud provides the value range of the 5-level standard of the occurrence possibility of all transitions. Through the cloud map of the judgment cloud, the failure possibility tendency of the transition can be judged under the ruler cloud standard. Combined with the operation process of the Petri net, the failure possibility of the transition is qualitatively judged under the 5-level standard of the ruler cloud, that is, the possibility level of the failure possibility of the transition in the failure language evaluation {extremely unlikely failure, less likely failure, generally possible failure, more likely failure, extremely likely failure} is judged. The judged failure possibility level and its corresponding interval value are used as the threshold value range for the activation of transition T in the operation of the Petri net. That is, a transition T i If the value in its judgment cloud belongs to the interval range of its scale cloud, then the transition T i After the transition is activated, the evolution shifts from the situation state represented by the input library to the situation state represented by the output library, thus realizing the failure possibility analysis from the situation judgment represented by the transition to the failure possibility analysis of the situation state represented by the output library.

[0070] In a specific example, firstly, based on the cloud forward generator, the scale cloud of the urban power grid system failure language evaluation is determined, and the digital characteristics of the five cloud models are determined according to the above formula. The calculation results are shown in Table 3. Table 3 is as follows:

[0071] Table 3 Language evaluation and cloud model of failure probability of power grid tripping cascading events

[0072]

[0073] The interval values ​​of the five cloud models are obtained according to the rule of the cloud model "3En", that is, more than 99.73% of the cloud droplets will fall on [E x -3E n , E x +3E n ], the interval values ​​of the five cloud models of the scale cloud for the risk language evaluation of power grid tripping chain events are determined. Through the forward cloud generator algorithm, the failure probability of the scale cloud of the power grid transmission line tripping secondary chain event transition is in the number domain interval [0, 1] with the cloud drop number N = 10000 drops, as shown in Figure 6 shown.

[0074] The transition failure probability analysis and evaluation of the secondary chain event transition of the power transmission line tripping based on the ruler cloud evaluates the language value level and the interval value corresponding to the five cloud models. Taking the evaluation unit P1-T1-P2 to which the transition T1 relay protection success belongs as an example, for the maximum and minimum values ​​of the probability of successful failure of the reclosing due to the success of the relay protection after the tripping corresponding to T1, the maximum and minimum cloud models are constructed through the reverse cloud generator algorithm. The digital characteristics are shown in Table 4. The calculation shows that the final judgment cloud model of T1 is C1 (0.5630, 0.0113, 0.0123).

[0075] Table 4 Maximum and minimum cloud models for transition T1

[0076]

[0077] Taking the T8 critical transmission line as an example, the number of cloud droplets N = 10,000 drops. Based on the failure probability cloud map of the scale cloud transition obtained by the forward cloud generator algorithm C8, the failure probability of the T8 critical transmission line is between the less likely failure and the generally possible failure in the scale cloud, and is more inclined to be less likely to fail.

[0078] In this embodiment, the risk scenario of secondary chain events caused by power transmission line tripping evolves from T1 to T 15 The numerical characteristics of the transition failure probability judgment cloud model are shown in Table 5. Table 5 is as follows:

[0079] Table 5 Digital features of the judgment cloud model of changes

[0080]

[0081] In this application, based on Monte Carlo simulation, according to the determined transition activation condition function, the improved Petri network is cyclically run until the number of operation calculations of the improved Petri network is equal to the number of Monte Carlo simulations Sim, and the ratio of the number of times the urban power grid system place events occur to the number of Monte Carlo simulations Sim is determined as the place failure probability. Among them, the place event represents the state change of the urban power grid system.

[0082] Specifically, first, initialize all variables of the improved Petri net, pre-set the number of Monte Carlo simulations Sim, define a counter i to record the number of Petri net runs, and define a variable to record the number of output places (i.e., the number of events triggered) after the transition is activated. Set the counter i of the number of runs of the improved Petri net = 0, and define P iTo record the output place after the transition is activated, i represents the variable of the number of times the event has occurred. Run the improved Petri network for calculation, starting from the initial node of the urban power grid accident from left to right, and the counter of the number of operations i = i + 1; then, clear the token value in the place after the previous round of network operation calculation, and restart the evolution of the initial node of the urban power grid. Therefore, only one token is placed in the initial node P1 place, indicating that only the P1 place event occurs, that is, the initial identifier M0 (1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0).

[0083] Next, the conditional function F of the activated transition is used to determine whether the transition is activated to form an evolutionary path. This requires external judgment cloud and ruler cloud data. The external judgment cloud data is used to simulate the possibility of each transition being activated, usually a random number or probability distribution; the external ruler cloud data defines numerical intervals of different probability levels, which are used to compare with the random numbers drawn from the judgment cloud to determine whether the transition is activated. Specifically, a set of random numbers is taken from the judgment cloud corresponding to the 15 transitions, and compared with the interval values ​​corresponding to the probability level of the ruler cloud. If the random number belongs to the probability level interval of the ruler cloud, the transition is activated. Among them, the interval values ​​corresponding to the probability level of the ruler cloud are:

[0084] {Very unlikely to failC -2 (E x-2 , E n-2 , H e-2 ), less likely to fail C -1 (E x-1 , E n-1 , H e-1 ), half of which may fail C0(E x0 , E n0 , H e0 ), more likely to fail C +1 (E x+1 , E n+1 , H e+1 ), Very likely to fail C +2 (E x+2 , E n+2 , H e+2 )}.

[0085] The expression of the condition function F for the transition to be activated is as follows:

[0086]

[0087] In the formula, C i For the change i The judgment cloud, normrnd(C i ) indicates that from the judgment cloud C i A random number taken fromi (max) indicates transition T i The maximum value of the failure probability level corresponding to the interval obtained based on the ruler cloud evaluation; B i (min) indicates transition T i The failure probability level obtained based on the scale cloud judgment corresponds to the minimum value of the interval. In a specific example, taking the critical transmission line of transition T8 as an example, its judgment cloud is C8 (0.3585, 0.0211, 0.0130), and the failure probability level of its scale cloud is less likely to fail, and the corresponding interval value is [0.117, 0.501]. The transition is activated when the condition function F for the transition to be activated is true, such as Figure 7 The figure shows the range of values ​​in which the conditional function F of transition T8 is true.

[0088] For a transition known to be activated by the conditional function F, it is necessary to determine whether there is a token in the corresponding input place to determine whether the event of the input place occurs, and to ensure the evolutionary continuity of the evolution path after starting from the initial node of place P1. If there is a token in the input place, then after the transition is activated, the number of tokens in the input place remains unchanged, and the number of tokens in the corresponding output place increases by 1. For example, in the evolution, only transition T1 is activated, and the initial identifier M0 (1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0) becomes M1 (1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0).

[0089] In a specific example, T1-T 15 From the judgment cloud, we can see that based on the 5-level standard of the scale cloud, the transition T1-T 15 The threshold selection based on the scale cloud is shown in Table 6, which is:

[0090] Table 6 Transition T1-T 15 Threshold based on ruler cloud

[0091]

[0092] After the improved Petri network is fully run once, the corresponding running results are recorded. The number of tokens in the identification vector M represents the evolution path of the urban power grid accident scenario in this operation. For the number of times the event represented by the output library has occurred, that is, P i =P i +M(i), where M(i) represents the value of the i-th element in the identification vector M, that is, the current number of tokens in the i-th library after the operation ends, reflecting the state change of the i-th library in this operation, and then revealing the dynamic behavior of the system represented by the entire improved Petri net model. The final result of the evolution of the secondary chain event risk scenario of the power grid line tripping is divided into a safe state and an unsafe state. The specific performance of the evolutionary final state is shown in Table 7. Table 7 is as follows:

[0093] Table 7 Final state of the evolution of the risk scenario of secondary chain events caused by power line tripping

[0094]

[0095] By judging whether the number of calculations i (i is a positive integer) of the improved Petri network is equal to the preset number of simulations Sim, if they are not equal, the new network operation will continue; if they are equal, the loop will be terminated and the probability of occurrence of the final state of evolution, that is, the probability of failure of the place, will be calculated. Among them, the probability of failure of the place of evolution for the final state of evolution is the ratio of the number of times the place event corresponding to the final state of evolution occurs to the number of Monte Carlo simulations Sim set. Among them, the place event represents the state change of the urban power grid system.

[0096] In the specific example of the evolution of the risk scenario of secondary chain events caused by power grid line tripping, the improved Petri network was run 10,000 times, but since there are still differences between different simulations, the average value of 5 simulations (each simulation contains 10,000 network cloud operation calculations) is used as the final result value. The final state probability value of the evolution of the risk scenario of secondary chain events caused by power grid line tripping is shown in Table 8. Table 8 is as follows:

[0097] Table 8 Final state probability values ​​of the risk scenario evolution of secondary chain events caused by power line tripping

[0098]

[0099]

[0100] By comparing the simulated evolutionary final state failure probability value with the scale cloud diagram and its qualitative value language level description, the final failure probability of the evolutionary final state can be obtained. Among them, the final failure probability of S1 recovery is a more likely failure, the final failure probability of S2 local recovery is a generally possible failure, the final failure probability of U1 transmission grid structure damage is a generally possible failure, and the final failure probability of U2 power grid blackout is a less likely failure.

[0101] Step S103: Determine the danger of the target storage place in the urban power grid system based on the cloud model.

[0102] In the present application, the target place is the place corresponding to the final state of the evolution of the event transition in the urban power grid system, and the danger of the target place is the failure probability corresponding to the final state place of the evolution path. Here, the danger of the target place is calculated based on the cloud model, which is similar to the process of determining the failure probability of the transition of the event transition in the urban power grid system based on the cloud model in step S102. First, based on the forward cloud generator, a scale cloud of the qualitative value language description corresponding to the danger of the target place of the urban power grid system is determined; then, based on the qualitative value language description of the danger of the target place defined by the scale cloud, a judgment cloud of the target place of the urban power grid system is generated based on the reverse cloud generator to determine the danger of the target place.

[0103] The difference is that the qualitative value language description (i.e., failure language evaluation) in step S102 is used to judge the failure probability of the urban power grid transition represented by the transition T, while the danger of the scenario state represented by the target library in the final state of the evolution is analyzed using different qualitative value language descriptions, and is also judged as 5 language degree levels, namely: {weak, minor, general, severe, catastrophic}.

[0104] In a specific example, taking the secondary chain event of power transmission line tripping as an example, first, a scale cloud of the degree of danger of the evolving final state in the dynamic risk assessment of the secondary chain event of power transmission line tripping is constructed, and a scale cloud diagram is output, as shown in Table 9 and Figure 8 shown.

[0105] Table 9 Language evaluation and cloud model of dangerous power grid tripping chain events

[0106]

[0107]

[0108] In this embodiment, P i The maximum and minimum possible intervals of the situational state danger represented by (i=3, 5, 8, 9, 11, 13, 15, 16) are judged to obtain the maximum cloud model C of the danger of the target library. max and minimum cloud model C min Then, the reverse cloud generator is used to process the cloud model of the severity of the evolutionary final state library event one by one, and the 8 evolutionary final state related libraries P are determined. i A cloud representing the severity of the situation Then, according to the formula of the comprehensive cloud model, the judgment cloud of the danger of the final state of the evolution of the risk scenario of the secondary chain event of the power transmission line tripping is established, as shown in Table 10:

[0109] Table 10 Digital features of the judgment cloud model of the evolutionary final state related library

[0110]

[0111] Next, the judgment cloud model of the evolutionary final states U1, U2, S1, and S2 is calculated, as shown in Table 11:

[0112] Table 11 Digital features of the final state judgment cloud model

[0113]

[0114] According to the cloud diagram of the hazard scale of the final state of evolution and its qualitative value language level description {weak, minor, general, severe, catastrophic}, the cloud diagram of the judgment of the hazard of the final state of evolution is shown in Table 12:

[0115] Table 12 Danger judgment cloud of the final state of evolution

[0116]

[0117] It can be seen that the risk of S1 recovery is weak, the risk of S2 partial recovery is general, the risk of U1 transmission grid structure damage is serious, and the risk of U2 power grid blackout is catastrophic.

[0118] Step S104: perform vulnerability assessment on the urban power grid system according to the power grid failure probability and the danger of the target storage location.

[0119] In this application, by assigning values ​​to the obtained power grid failure probability and dangerous language degree level, an urban power grid vulnerability evaluation matrix is ​​established to evaluate the vulnerability level of each library. Specifically, the product of the power grid failure probability and the dangerousness of the target library is calculated to determine the vulnerability of the urban power grid system; then, the qualitative value language description (i.e., failure language level) of the power grid failure probability and the dangerousness of the target library is assigned respectively to construct a vulnerability assessment matrix for the urban power grid system. Among them, the failure language level characterizes the probability of the evolution of urban power grid accidents. Finally, based on the vulnerability assessment matrix, the vulnerability level of the urban power grid system is determined according to the vulnerability of the urban power grid system.

[0120] Here, five different language levels are assigned for the probability of power grid failure and the dangerousness of the target library, where the power grid failure probability score (extremely unlikely failure = 1, less likely failure = 2, moderately likely failure = 3, more likely failure = 4, extremely likely failure = 5); the dangerousness score of the target library (weak = 1, minor = 2, general = 3, severe = 4, catastrophic = 5).

[0121] In the established vulnerability assessment matrix, the rows of the matrix represent the probability of grid failure, and the columns represent the danger of the target storage location. Each cell of the vulnerability matrix is ​​the product of two scores, representing the vulnerability score of a specific combination. Finally, different vulnerability levels are divided according to the range of vulnerability scores. Among them, 1-5: very low vulnerability; 6-10: low vulnerability; 11-15: medium vulnerability; 16-20: high vulnerability; 21-25: very high vulnerability.

[0122] The final vulnerability assessment result can be obtained according to the language level assignment and vulnerability assessment matrix. In the specific example of the secondary chain event of power transmission line tripping, the vulnerability level of S1 is very low vulnerability, the vulnerability level of S2 is low vulnerability, the vulnerability level of U1 is medium vulnerability, and the vulnerability level of U2 is low vulnerability. Finally, based on the vulnerability assessment results, corresponding mitigation measures are given.

[0123] In this embodiment, since the vulnerability level of S1 is very low, it is only necessary to maintain the status quo and conduct routine inspections, regularly monitor and maintain to ensure system stability, and implement regular risk assessments to prevent potential small changes; the vulnerability level of S2 is low, but consideration should be given to enhancing local recovery capabilities, such as by increasing backup resources or optimizing the recovery process, and strengthening monitoring of key areas, and formulating a rapid response mechanism to reduce potential impacts. U1: The vulnerability level of structural damage to the transmission grid is medium vulnerability, and it is necessary to focus on the stability of U1, strengthen the physical structure of the transmission grid, improve its disaster resistance, such as earthquake resistance and flood prevention measures, and regularly conduct structural inspections and maintenance to repair potential weak links in a timely manner; U1: The vulnerability level of a large power outage in the power grid is low vulnerability, but due to the extremely high risk of a large power outage, a strict emergency plan should be formulated to ensure rapid recovery capabilities, and redundant systems and backup power supply lines should be added to improve the overall resilience and fault tolerance of the system.

[0124] The present application also provides a power grid failure probability and vulnerability assessment system based on an improved Petri network, such as Fig. 9 As shown in the figure, the assessment system is used to assess the failure probability and vulnerability of the urban power grid system, including:

[0125] A Petri network improvement unit 901 is configured to improve the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri network to obtain an improved Petri network of the urban power grid system;

[0126] A failure judgment unit 902 is configured to determine a power grid failure probability of the urban power grid system based on an improved Petri network of the urban power grid system;

[0127] Danger determination 903 is configured to determine the danger of a target place in the urban power grid system based on a cloud model; wherein the target place is a place corresponding to the final state of the evolution of the event transition in the urban power grid system;

[0128] The vulnerability unit 904 is configured to perform vulnerability assessment on the urban power grid system according to the probability of power grid failure and the danger of the target storage place.

[0129] The power grid failure probability and vulnerability assessment system based on the improved Petri network provided in the embodiment of the present application can implement the steps and processes of the power grid failure probability and vulnerability assessment method based on the improved Petri network described in any of the above embodiments, and achieve the same technical effects, which will not be repeated here one by one.

[0130] In the description of the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0131] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for evaluating the failure probability and vulnerability of a power grid based on an improved Petri network, which is used to evaluate the failure probability and vulnerability of a city power grid system, and is characterized in that: include: Step S101, improving the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri network to obtain an improved Petri network of the urban power grid system; Step S102: determining a power grid failure probability of the urban power grid system based on an improved Petri network of the urban power grid system; Step S103: determining the danger of a target place in the urban power grid system based on the cloud model; wherein the target place is a place corresponding to the final state of the evolution of the event transition in the urban power grid system; Step S104: performing a vulnerability assessment on the urban power grid system according to the power grid failure probability and the danger level of the target storage location.

2. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 1 is characterized in that: In step S101, the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri net are improved, including: The input library P of the urban power grid system i The input correlation matrix of the transition T and the output library P of the transition T to the urban power grid system j The output association matrix is ​​changed; wherein, the input association matrix is ​​allowed to have at most one arc from the input library P i To the transition T; the output association matrix allows at most one arc from the transition T to the output library P j ; The input association matrix represents the input library P i The output correlation matrix represents the input relationship from the transition T to the output library P. j The output relationship of; i, j are both positive integers; Determine that the event identifier M of each place P of the urban power grid system is: P→{0, 1}, indicating the event state of the place P, wherein 0 indicates that the event corresponding to the place P has not occurred, and 1 indicates that the event corresponding to the place P has occurred; Determine that the preset transition activation threshold is met in the urban power grid system, and the input library P i The corresponding event has occurred, the transition T is activated, and the input library P i The event state evolves to the output library P j The event status.

3. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 1 is characterized in that: In step S101, According to the historical fault data and operation data of the urban power grid system, the key factors affecting the evolution of failure events of the urban power grid system are extracted and searched based on the Delphi method to construct an improved Petri network of the urban power grid system.

4. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 1 is characterized in that: Step S102 includes: Based on the cloud model, determining the transition failure probability of the event transition in the urban power grid system; wherein the event transition represents the evolution of the power grid accident of the urban power grid system; Based on the improved Petri net of the urban power grid system, the failure probability of the storage place of the urban power grid system is determined by Monte Carlo simulation; Based on the transition failure probability and the library failure probability, a power grid failure probability of the urban power grid system is determined.

5. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 4 is characterized in that: The determining, based on the cloud model, the transition failure probability of the event transition in the urban power grid system includes: Based on the forward cloud generator, a scale cloud of the failure language evaluation of the urban power grid system is determined; wherein the failure language evaluation represents the failure probability of the event transition; According to the failure language evaluation of the event transition defined by the ruler cloud, based on the reverse cloud generator, a judgment cloud of the event transition in the urban power grid system is generated to determine the transition activation threshold of the event transition, and whether the event transition in the urban power grid system is valid is judged based on the transition activation threshold.

6. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 5 is characterized in that: The method of determining the scale cloud for the urban power grid system failure language evaluation based on the forward cloud generator includes: Based on the cloud model "3En" rule, the interval values ​​of multiple cloud models of the scale cloud of the urban power grid system failure language evaluation are determined, and the scale cloud cloud diagram of the failure probability of the event transition is calculated through the forward cloud generator algorithm.

7. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 5 is characterized in that: The improved Petri net based on the urban power grid system determines the failure probability of the storage place of the urban power grid system through Monte Carlo simulation, including: Based on Monte Carlo simulation, according to the determined transition activation condition function, the improved Petri network is run in a loop until the number of running calculations of the improved Petri network is equal to the set number of Monte Carlo simulations Sim, and the ratio of the number of times the urban power grid system place event occurs to the set number of Monte Carlo simulations Sim is determined as the place failure probability; wherein the place event represents the state change of the urban power grid system.

8. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 1 is characterized in that: In step S103, The hazard of the target library is determined according to the pre-set qualitative value language description and the hazard scale cloud diagram of the final state of evolution.

9. The urban power grid failure probability analysis and vulnerability assessment method based on improved Petri network according to claim 1 is characterized in that: In step S104, Calculate the product of the power grid failure probability and the danger of the target storage location to determine the vulnerability of the urban power grid system; Assigning values ​​to the failure probability of the power grid and the failure language level of the dangerousness of the target library respectively to construct a vulnerability assessment matrix of the urban power grid system; wherein the failure language level represents the probability of the evolution of the urban power grid accident; Based on the vulnerability assessment matrix and according to the vulnerability of the urban power grid system, a vulnerability level of the urban power grid system is determined.

10. A power grid failure probability and vulnerability assessment system based on an improved Petri network, used for evaluating the failure probability and vulnerability of an urban power grid system, characterized in that: include: A Petri network improvement unit configured to improve the input relationship, output relationship, event identifier M, transition activation and execution rules in the traditional Petri network to obtain an improved Petri network of the urban power grid system; a failure judgment unit configured to determine a power grid failure probability of the urban power grid system based on an improved Petri network of the urban power grid system; Danger judgment is configured to determine the danger of a target library in the urban power grid system based on a cloud model; wherein the target library is a library corresponding to the final state of the evolution of the event transition in the urban power grid system; The vulnerability unit is configured to perform a vulnerability assessment on the urban power grid system according to the power grid failure probability and the danger of the target storage location.

Images