Power distribution network line reinforcement decision-making method and system considering typhoon disaster

Abstract

The invention discloses a power distribution network line reinforcement decision-making method considering a typhoon disaster, and the method comprises the steps: generating the fault probability of each line after the typhoon disaster according to a power distribution network pole fault probability generation method, and obtaining the fault state of each line in a power distribution network after the typhoon disaster, the load data of a non-electric vehicle user, and the initial state of an electric vehicle; then establishing a post-disaster flexible resource cooperation rapid recovery model, simulating a recovery process after a typhoon disaster according to the post-disaster flexible resource cooperation rapid recovery model to obtain a repair moment of each line, and performing sorting by using a Scioplande sorting method according to the repair moment of each line so as to obtain a line importance degree sequence; and formulating a line typhoon prevention reinforcement decision of the power distribution network according to the line importance degree sequence. The method can be used for making the line reinforcement strategy of the power distribution network under the extreme natural disaster of the typhoon.

Description

Technical Field

[0001] The present invention relates to the field of power system safety planning and operation, and particularly to a decision-making method and system for reinforcing distribution network lines under typhoon disasters. Background Art

[0002] In recent years, extreme natural disasters have occurred frequently. Natural disasters often cause large-scale power outages in the distribution network, resulting in huge economic and social losses. It is urgent to study the reinforcement strategies for the distribution network to cope with extreme natural disasters. Quick and efficient restoration is the key step to improve the resilience of the power distribution system. Scalable and flexible resources have become the construction direction and trend of the new power system. Utilizing flexible resources and realizing the coordinated scheduling among different types of flexible resources contribute to the rapid restoration of the distribution network after disasters. Flexible resources such as mobile energy storage, distributed power sources, and electric vehicles are becoming increasingly popular. By reconstructing the transmission path, forming a microgrid to supply power to the region, optimizing the scheduling method of elastic resources, and using mobile energy storage, vehicle grid-connected power stations, and distributed power sources to supply power to the distribution network, a self-sufficient microgrid can be formed, which can effectively improve the restoration ability of the post-disaster system whether it operates in island mode or grid-connected mode. At the planning level, anti-typhoon reinforcement measures are formulated for the distribution network, but existing research rarely considers the coordinated scheduling of flexible resources after disasters. Summary of the Invention

[0003] The purpose of the present invention is to provide a decision-making method and system for reinforcing distribution network lines under typhoon disasters in view of the defects of the existing technology.

[0004] To achieve this purpose, a decision-making method for reinforcing distribution network lines under typhoon disasters designed by the present invention includes the following steps:

[0005] Generating the failure probability of each line after the occurrence of a typhoon disaster according to the failure probability generation method of distribution network poles;

[0006] Obtaining the failure states of each line in the distribution network after the occurrence of a typhoon disaster by using non-sequential Monte Carlo simulation according to the failure probability of each line after the occurrence of a typhoon disaster;

[0007] Generating the non-electric vehicle user load data after the occurrence of a typhoon disaster according to the distribution network topology data and the normal load data of non-electric vehicle users;

[0008] Generating the initial state of electric vehicles after the occurrence of a typhoon disaster according to the electric vehicle parameters;

[0009] Establishing a post-disaster flexible resource collaborative rapid restoration model according to the failure states of each line in the distribution network after the occurrence of a typhoon disaster, the non-electric vehicle user load data after the occurrence of a typhoon disaster, the initial state of electric vehicles after the occurrence of a typhoon disaster, the objective function of the post-disaster restoration process, and the constraint conditions of the distribution network fault scenarios;

[0010] Simulate the recovery process after a typhoon disaster according to the post-disaster flexible resource collaborative rapid recovery model to obtain the repair times of each line;

[0011] Sort according to the repair times of each line using the Copeland ranking method to obtain the line importance ranking;

[0012] Formulate the line typhoon protection reinforcement decision of the distribution network according to the line importance ranking.

[0013] A distribution network line reinforcement decision-making system considering typhoon disasters, which includes a line fault probability generation module, a fault state acquisition module, a load data generation module, an electric vehicle initial module, a recovery model establishment module, a repair time calculation module, a Copeland ranking module and a decision-making module;

[0014] The line fault probability generation module is used to generate the fault probabilities of each line after a typhoon disaster according to the distribution network pole fault probability generation method;

[0015] The fault state acquisition module is used to obtain the fault states of each line in the distribution network after a typhoon disaster by using non-sequential Monte Carlo simulation according to the fault probabilities of each line after a typhoon disaster;

[0016] The post-disaster data generation module is used to generate non-electric vehicle user load data and electric vehicle initial states after a typhoon disaster according to the fault probabilities of each line after a typhoon disaster;

[0017] The load data generation module is used to generate non-electric vehicle user load data after a typhoon disaster according to the distribution network topology data and non-electric vehicle user normal load data;

[0018] The electric vehicle initial module is used to generate the initial states of electric vehicles after a typhoon disaster according to the electric vehicle parameters;

[0019] The recovery model establishment module is used to establish a post-disaster flexible resource collaborative rapid recovery model according to the fault states of each line in the distribution network after a typhoon disaster, the non-electric vehicle user load data after a typhoon disaster, the initial states of electric vehicles after a typhoon disaster, the objective function of the post-disaster recovery process and the constraint conditions of the distribution network fault scenario;

[0020] The repair time calculation module is used to simulate the recovery process after a typhoon disaster according to the post-disaster flexible resource collaborative rapid recovery model to obtain the repair times of each line;

[0021] The Copeland ranking module is used to sort according to the repair times of each line using the Copeland ranking method to obtain the line importance ranking;

[0022] The decision-making module is used to formulate the typhoon-proof reinforcement decision of the distribution network according to the sorting of line importance.

[0023] Advantages of the present invention:

[0024] The present invention proposes a method for making reinforcement decisions for distribution network lines considering typhoon disasters. The present invention considers that after natural typhoon disasters such as typhoons occur, the distribution network uses a variety of flexible resources to cooperate and quickly recover, and formulates typhoon-proof reinforcement strategies for the distribution network according to the Copeland sorting results. The Copeland sorting method described in the present invention can consider the repair order of limited personnel / resources, and other reinforcement decisions cannot consider the repair order under uncertain fault sets, such as stochastic programming; the method described in the present invention considers the rapid recovery of a variety of flexible resources after the disaster, and the formulated line reinforcement decisions are more in line with the actual situation. This method can provide a scientific reference basis for the system typhoon-proof reinforcement planning of the distribution network in response to extreme events. Description of the drawings

[0025] Figure 1 It is the program flow chart of the present invention.

[0026] Figure 2 It is a schematic diagram of the objective function of the flexible resource collaborative rapid recovery model for the distribution network after the disaster.

[0027] Figure 3 It is a schematic diagram of the subgraph formed during the network reconstruction process.

[0028] Figure 4 It is the structural schematic diagram of the present invention. Specific implementation manners

[0029] The following further detailed description of the present invention is provided in conjunction with the drawings and specific embodiments:

[0030] As Figure 1 shown, a method for making reinforcement decisions for distribution network lines considering typhoon disasters includes the following steps;

[0031] Generate the failure probability of each line after the typhoon disaster according to the distribution network pole failure probability generation method;

[0032] According to the failure probability of each line after the typhoon disaster, use non-sequential Monte Carlo simulation to obtain the failure state of each line in the distribution network after the typhoon disaster;

[0033] Generate the non-electric vehicle user load data after the typhoon disaster according to the distribution network topology data and the normal load data of non-electric vehicle users;

[0034] Generate the initial state of electric vehicles (power consumption load and initial position of electric vehicles) after the typhoon disaster according to the electric vehicle parameters;

[0035] According to the fault states of each line in the distribution network after the typhoon disaster, the non-electric vehicle user load data after the typhoon disaster, the initial state of electric vehicles after the typhoon disaster, the objective function of the post-disaster recovery process, and the constraint conditions of the distribution network fault scenario, a collaborative and rapid recovery model of flexible resources after the disaster is established;

[0036] According to the collaborative and rapid recovery model of flexible resources after the disaster, simulate the recovery process after the typhoon disaster to obtain the repair time of each line;

[0037] Sort according to the repair time of each line using the Copeland ranking method to obtain the line importance ranking;

[0038] Formulate the line typhoon prevention and reinforcement decision of the distribution network according to the line importance ranking.

[0039] A method for making line reinforcement decisions in a distribution network considering typhoon disasters. The specific method for generating the fault probability of each line according to the distribution network pole fault probability generation method is as follows:

[0040] The fault probability caused by the impact of the typhoon on the pole:

[0041]

[0042] Where a' is a pole on line ij; w(t) represents the wind speed at time t is the fault probability of pole a'; μ is the resistance of the pole, ξ is the logarithmic standard deviation of typhoon intensity measurement; p ij is the fault probability of line ij; m is the number of poles on line ij.

[0043] A method for making line reinforcement decisions in a distribution network considering typhoon disasters. The specific method for obtaining the fault states of each line in the distribution network after the typhoon disaster using non-sequential Monte Carlo simulation according to the fault probabilities of each line after the typhoon disaster is as follows:

[0044] Extract the states of each line of the system through the non-sequential Monte Carlo method, and extract a random number rand i , this random number is uniformly distributed in the [0,1] space, p ij is the fault probability of line ij, and obtain the operating state s i (0) of line ij at the start of the repair process after the typhoon disaster:

[0045]

[0046] A method for making line reinforcement decisions in a distribution network considering typhoon disasters. The specific method for generating the non-electric vehicle user load data after the typhoon disaster according to the distribution network topology data and the non-electric vehicle user normal load data is as follows:

[0047] Non - electric vehicle user load data generation:

[0048] To be closer to the real typhoon disaster, sample and generate the non - electric vehicle user load after the typhoon disaster occurs; assume that the load of power system node i is the product of the first random multiplier and the second random multiplier and the hourly standardized load curve:

[0049]

[0050] where Ν represents the set of power system nodes; and represents the average value of the daily peak active power of power system node i in a year, represents the average value of the daily peak reactive power of power system node i in a year.

[0051]

[0052] where Τ is the set of time; P i L (t) is the active load of the non - electric vehicle load of power system node i at time t, is the reactive load of the non - electric vehicle load of power system node i at time t; M p (t) is the standardized reactive load curve varying with time, M q (t) is the standardized reactive load curve varying with time; since the typhoon may occur at any time, it is assumed that the time offset of the load curve is uniformly distributed;

[0053] A decision - making method for strengthening the distribution network lines considering typhoon disasters. According to the electric vehicle parameters, the specific method for generating the initial state of electric vehicles after the typhoon disaster occurs is as follows:

[0054] Initial state generation of electric vehicles:

[0055] The present invention considers that during typhoon disasters, some electric vehicles participate in the restoration of the distribution network and supply power to the grid in response to policies, and the rest of the electric vehicles are regarded as the load part of ordinary load users. The electric vehicle parameters include the initial position of the electric vehicle, the time when the electric vehicle enters the charging station, the daily driving mileage of the electric vehicle, and the battery capacity of the electric vehicle; by sampling the initial position, the time of entering the charging station, the daily driving mileage, and the battery capacity of the electric vehicle, the initial state of the electric vehicle cluster is obtained;

[0056] The initial position of the electric vehicle is according to the distance to the nearest V2G (vehicle to grid) vehicle grid - connected charging station O tIs uniformly distributed;

[0057] The time when an electric vehicle enters the charging station follows a normal distribution, and its probability distribution is:

[0058]

[0059] where t ch is the time when the electric vehicle enters the charging station; μ ch is the mean value of the time to enter the charging station; σ ch is the standard deviation of the time to enter the charging station;

[0060] The daily driving mileage of an electric vehicle follows a lognormal distribution, and its probability distribution is:

[0061]

[0062] where s is the daily driving mileage; μ s is the mean value of the daily driving mileage; σ s is the standard deviation of the daily driving mileage;

[0063] The battery capacities of different types of electric vehicles follow a normal distribution, and its probability distribution is:

[0064]

[0065] where C p is the battery capacity of the electric vehicle; μ c is the mean value of the battery capacity; σ c is the standard deviation of the battery capacity;

[0066] The specific steps to obtain the electric vehicle user load data in the scenario are as follows: First, extract the battery capacity of a single electric vehicle according to the probability distribution of the electric vehicle battery capacity; extract the driving mileage according to the probability distribution of the daily driving mileage, calculate the consumed battery power, and obtain the SOC state (state of charge) of the electric vehicle battery when entering the charging station; extract the time to enter the charging station according to the probability distribution of the time to enter the charging station; extract the charging station where the electric vehicle charges according to the probability distribution of the electric vehicle location, and loop through all electric vehicles to calculate the electric vehicle user load P i L,E (t);

[0067] The active power load of power system node i at time t is the sum of the electric vehicle user load and the non-electric vehicle user load, that is, P L,i,t =P i L (t)+P i L,E (t).

[0068] A decision-making method for strengthening distribution network lines under typhoon disasters. The specific method for establishing a collaborative and rapid recovery model of flexible resources after the disaster is based on the fault status of the lines, non-electric vehicle user load data, initial state of electric vehicles, objective function of the post-disaster recovery process, and constraint conditions of the distribution network fault scenarios. It is characterized by further including the following steps:

[0069] Considering the time for personnel movement, determine the repair order of the lines during the recovery process to maximize the resilience of the recovery process, and identify the lines that need to be strengthened according to the line recovery order; the model consists of the objective function of the post-disaster recovery process and the constraint conditions of the distribution network fault scenarios. The constraint conditions of the distribution network fault scenarios include line working state constraints, distribution network power flow constraints, radial state topology constraints, fault impact transfer constraints under N-k faults of the distribution network, mobile energy storage constraints, demand response constraints, electric vehicle constraints, and repair personnel path constraints. This invention needs to satisfy all the constraint conditions simultaneously;

[0070] This step is to obtain the repair time of each component after the extreme event occurs. It is necessary to establish a model that simulates the post-disaster recovery process considering the collaborative scheduling of flexible resources and output the recovery time of each component in a specific scenario.

[0071] The entire process of encountering an extreme typhoon disaster can be represented by Figure 2 This method determines the repair order of the lines during the recovery process considering the time for personnel movement to maximize the resilience of the recovery process, and identifies the components that need to be strengthened according to this recovery order.

[0072] The goal of establishing a collaborative and rapid recovery model of flexible resources after the disaster is to minimize the load shedding during the entire post-disaster recovery process. The objective function of the post-disaster recovery process is:

[0073]

[0074] where represents the weight of power system node i; P shed,i,t represents the active load shed at power system node i at time t; where, Δt is the time step, assuming Δt is 15 minutes;

[0075] Line working state constraints:

[0076]

[0077] where s ij (t) is a 0-1 variable. When the ij line fails, s ij (t) = 0; when it is working normally, s ij (t) = 1; (i,j) ∈ E' is the set of fault lines; (i,j) ∈ E is the set of lines;

[0078] Distribution network power flow constraint:

[0079] Use the linearized DistFlow model to construct operation constraints; among them, the power balance constraint of each power system node is:

[0080]

[0081] Where Ν represents the set of power system nodes; the node in the inflow direction of power system node i is the parent node of power system node i, and λ(i) is defined as the set of parent nodes of power system node i; the node in the outflow direction of power system node i is the child node of power system node i, and δ(i) is defined as the set of child nodes of power system node i; P ij,t The active power flowing through line (i,j) at time t; Q ij,t Represents the reactive power flowing through line (i,j) at time t; P L,i,t Represents the active load of power system node i at time t; Q L,i,t Represents the reactive load of power system node i at time t; P DG,i,t Represents the active power output of DG at power system node i at time t; Q DG,i,t Represents the reactive power output of DG at power system node i at time t; P shed,i,t Represents the active load cut-off at power system node i at time t; Q shed,i,t Represents the reactive load cut-off at power system node i at time t;

[0082] If the branch is closed, the voltage difference of the branch is restricted by the power flow. If the branch is open, the voltage difference is arbitrary and the branch flow must be zero. The voltage relationship constraint between adjacent nodes is:

[0083]

[0084] Where U i,t Represents the voltage at power system node i at time t, U j,t Represents the voltage at power system node j at time t; U0 represents the rated voltage; r ij And x ij Represent the resistance and reactance of line (i,j); c ij,t Is a 0-1 variable of the state of line (i,j) at time t, 0 means open, 1 means closed; M represents the first constant, and the value range of M is [10 3 , +∞], and in the present invention, 10 3 ;

[0085] The capacity limit constraint of each line is:

[0086]

[0087] wherein represents the maximum allowable power flowing through line (i, j);

[0088] The maximum and minimum voltage constraints of each power system node are:

[0089]

[0090] where Ν\{DG} represents the set of nodes without installed distributed power sources; and are the maximum and minimum voltage limits at power system node i;

[0091] The output limit constraint of the distributed generator (DG) is:

[0092]

[0093] where {DG} represents the set of nodes with installed distributed power sources; and is the minimum active power limit of DG at power system node i; is the maximum active power limit of DG at power system node i; is the minimum active power limit of DG at power system node i; is the maximum active power limit of DG at power system node i; represents a 0-1 variable indicating whether power system node i is affected by a fault. When power system node is affected by a fault takes 1, otherwise takes 0;

[0094] The demand limit constraint of the controllable load is:

[0095]

[0096] Radial state topology constraint:

[0097] During the network reconstruction and recovery process after a fault, the subgraph may include a subgraph connecting the substation, a subgraph powered only by DG, and a passive island subgraph. As Figure 3 shown, after lines 1-2 and 8-9 are attacked, passive island subgraphs are formed at nodes 2-3 and 8. At this time, the tie line is closed, and nodes 4, 5, 9, and 10 form a subgraph powered by DG, and the remaining nodes form a subgraph connecting the substation. Through such a distribution network reconstruction, flexible reconstruction zoning can be achieved, which not only supports the formation of a microgrid powered by distributed power sources but also supports the formation of load islands.

[0098] The number of closed lines is equal to the number of nodes Nnode Subtracting 1, and then subtracting the number of subgraphs powered only by DG and the number of islanded subgraphs, the necessary and sufficient condition 1 for a radial topology can be satisfied. The relevant constraints are:

[0099]

[0100] where N node represents the number of nodes; N sub represents the number of substations; γ i,t is a 0-1 variable indicating whether a node is a source node;

[0101] The variable γ i,t is only allowed to take 1 when the line connected to power system node i is disconnected. The relevant constraints are:

[0102]

[0103] In each subgraph, the virtual load can obtain the power supply from the virtual power source, that is, each subgraph is a connected graph. The relevant constraints are:

[0104]

[0105] where {Sub} represents the set of substation nodes; N\{Sub} represents the set of nodes excluding substation nodes. N node represents the number of nodes; N sub represents the number of substations; γ i,t is a 0-1 variable indicating whether a node is a source node; V ij,t represents the virtual power of line (i,j) at time t; W i,t represents the output of the virtual power source at power system node i, which is an unrestricted real number. In the virtual network, substation nodes and nodes with γ i,t = 1 are regarded as source nodes, and an unrestricted virtual power source output power is set. For other nodes, a virtual load with a load value of 1 is set.

[0106] Fault impact transfer constraints under N-k faults in the distribution network

[0107]

[0108] where ε is the second constant, and the value range of ε is [0, 0.01]. In the present invention, 0.01 is taken; y ij1,t is a binary variable representing the switch state of the i-side of line (i,j) at time t, and y ij2t is a binary variable representing the switch state of the j-side of line (i,j) at time t. If no switch is configured or the switch is in the closed state, it takes 1, otherwise it takes 0; is a binary parameter. If the i-side (κ = 1) or j-side (κ = 2) of line (i,j) is equipped with RCS, it takes 1, otherwise it takes 0; A 0-1 variable indicating whether node i is affected by a fault. When the node is affected by a fault it takes the value 1, otherwise it takes the value 0;

[0109] Mobile energy storage constraint:

[0110]

[0111] where {MESS} is the set of mobile energy storages; {dep} is the set of warehouses from which the mobile energy storages depart; N MESS is the total number of mobile energy storages; d i,e,t is a 0-1 variable indicating whether mobile energy storage e is at power system node i at time t; is the routing time for mobile energy storage to move from power system node i to power system node j;

[0112] Equation (26) indicates that a mobile energy storage can be connected to at most one power system node per time period; Equation (27) shows that a mobile energy storage can provide power to power system node i only after it is connected to power system node i; Equation (28) means that a mobile energy storage does not change its position after being connected to a power system node;

[0113]

[0114]

[0115] where is the upper limit of the state of charge of mobile energy storage e, is the lower limit of the state of charge of mobile energy storage e; E e,t is the state of charge of mobile energy storage e at time t; σ e is the self-discharge rate of mobile energy storage e, η e is the charge-discharge efficiency of mobile energy storage e; is the active power output of mobile energy storage e at power system node i at time t, is the reactive power output of mobile energy storage e at power system node i at time t; represents the total power output of mobile energy storage at power system node i at time t; is a 0-1 variable indicating the working state of mobile energy storage e at node i at time t; represents the minimum active power of mobile energy storage e, represents the minimum reactive power of mobile energy storage e; represents the maximum active power of mobile energy storage e, represents the maximum reactive power of mobile energy storage e;

[0116] Demand response constraint:

[0117]

[0118] Among them is the demand response load participated by power system node i at time t; P L,i,t represents the active power load of power system node i at time t; P i max represents the maximum limit of the active power load of power system node i; is the load of power system node i after participating in demand response at time t; is a 0-1 variable indicating whether power system node i is equipped with a demand response device; α i,t is the participation ratio of demand response load at time t; P i LD,max is the maximum allowable load reduction of power system node i;

[0119] Electric vehicle constraint:

[0120] Under extreme natural disasters, the travel willingness of electric vehicle (EV) users drops significantly. EV stations have good shelter effects. The government can promote and guide EVs to go to the stations and participate in reverse power transmission to the power grid. The operation characteristics of EVs are described using the travel chain theory. The set expression of the travel chain is:

[0121] L = {B0, B f , W 0f , L 0f , T0, T f , T 0f , T p} (38)

[0122] Among them, L is the travel chain set; the set elements are the travel chain starting point B0, the travel chain ending point B f , the travel path W 0f , the corresponding driving distance L 0f , the departure time T0, the arrival time T f , the driving time T 0f and the parking time T p ;

[0123] The time constraint for an electric vehicle to reach the vehicle to grid (V2G) station is as follows:

[0124]

[0125] Among them, t 0,n represents the departure time of the nth section of the journey; represents the arrival time of the nth section of the journey; represents the driving time of the nth section of the journey; is the residence time of the nth section of the journey;

[0126] Electric vehicle travel willingness constraint:

[0127]

[0128] Among them is the number of EVs participating in the dispatching response; β is the proportion of EVs participating in the restoration; N EV is the total number of EVs;

[0129] Electric vehicle real-time status constraint:

[0130]

[0131] Among them is the set of real-time information of each EV at time t, including the remaining power of the current EV driving status w t and the number O of the V2G station closest to the current location t ;

[0132] Electric vehicle cluster state of charge constraint:

[0133]

[0134] Among them is the state of charge of electric vehicle k at time t; is the charging and discharging efficiency of electric vehicle k; is the output of electric vehicle k at power system node i; is a 0-1 variable indicating whether electric vehicle k is at power system node i; {EV} represents the set of electric vehicles; Δt is the time step;

[0135] V2G reverse charging output constraint:

[0136]

[0137] Among them P i V2G,max 、P i V2G,min are the maximum and minimum outputs of the V2G station located at power system node i, respectively; is the output of the V2G station located at power system node i in time period t; is the maximum number of vehicles that the V2G station located at power system node i can accommodate; V G represents the set of V2G stations;

[0138] Maintenance personnel path constraint:

[0139] Ensure that the maintenance team starts from the specified departure location, and the constraint is:

[0140]

[0141] Among them, Crew represents the set of maintenance teams; {dep} represents the set of departure positions of the maintenance teams; represents the time when maintenance team c arrives at line m;

[0142] The line can only be repaired after the team arrives, so the relevant constraint is:

[0143]

[0144] Among them, y m,c represents whether maintenance team c arrives at line m; f m,t represents whether line m is repaired at time t. If it is, it is equal to 1, otherwise it is equal to 0;

[0145] The time when the maintenance team arrives at the line and f m,t The relationship is:

[0146]

[0147] Among them represents the time required for maintenance team c to repair line m;

[0148] Used to couple f m,t with the line status s m (t) The constraint is:

[0149]

[0150] All maintenance teams can only go to the faulty line once and can only leave once. The relevant constraints are as follows:

[0151]

[0152] Among them, Crew represents the set of maintenance teams; {dep} represents the set of departure positions of the maintenance teams; E'∪{dep}\{n} represents the set of damaged lines and warehouses excluding line n; x m,n,c represents a 0-1 variable indicating whether maintenance team c moves from line m to line n. If it is, it is equal to 1, otherwise it is equal to 0; represents the time consumed for maintenance team c to move from line m to line n;

[0153] Use the big M method to make the route of the maintenance personnel continuous. The relevant constraints are as follows:

[0154]

[0155] Among them represents the time consumed for maintenance team c to move from line m to line n; Denote the time when repair team c arrives at line m; Denote the time when repair team c arrives at line n; Denote the time required for repair team c to repair line m; Denote the time consumed for repair team c to move from line m to line n; y m,c Denote whether repair team c arrives at line m;

[0156] A decision-making method for strengthening distribution network lines under typhoon disasters, which simulates the recovery process after typhoon disasters according to the post-disaster flexible resource collaborative rapid recovery model. The specific method for obtaining the repair time of each line is as follows:

[0157] The repair time of each line is calculated by the time when the maintenance personnel arrive at the line and the maintenance duration of the line:

[0158]

[0159] Where T m Denote the time when line m is repaired and completed.

[0160] A mixed integer linear programming model is formed by the objective function and constraint conditions. After solving, the repair time of each line can be obtained.

[0161] A decision-making method for strengthening distribution network lines under typhoon disasters, which uses the Copeland ranking method to rank according to the repair time of each line. The specific method for obtaining the line importance ranking is as follows:

[0162] According to the above Monte Carlo sampling method, multiple groups of fault scenarios can be obtained by sampling the fault scenarios; solving the above model for these fault scenarios can obtain the repair time of each line in each scenario and form the distribution function of the repair time of each line; the comparison of the distribution functions can be carried out by the Copeland ranking method; compared with the ordinary planning method, the line importance ranking obtained by the Copeland ranking, because this method can compare multiple characteristic quantities of the line data samples, can consider more elements such as post-disaster maintenance and power grid reconstruction, and finally is expressed by the repair time, which is more conducive to post-disaster recovery and provides decision-making support for the strengthening of poles on important lines of the system and the formulation of system recovery strategies;

[0163] The Copeland score S of line m m,k The calculation method is as follows:

[0164]

[0165] Where q k (m) represents the k-th percentile of the cumulative distribution function of the repair time of line m; S m,n,kDenote the Copeland score after the k - th comparison between line m and line n; S m Denote the total Copeland score of line m;

[0166] First, perform Ω pairwise Copeland comparisons between line m and all other lines to obtain S m,n,k ; then sum up the Copeland scores S m,n,k of each pairwise comparison to obtain the final Copeland score S m of line m.

[0167] Rank the importance of lines according to the Copeland score S m of the lines.

[0168] A decision - making method for strengthening distribution network lines considering typhoon disasters. The specific method for formulating the typhoon - proof strengthening decision of the distribution network lines according to the ranking of line importance is as follows:

[0169] Rank according to the line importance, identify the lines with top rankings, and use the supporting poles on these lines as the distribution network poles to be strengthened. The number of reinforcements is determined according to the reinforcement budget:

[0170]

[0171] Where Denote the cost of reinforcing all poles supporting line ij; Denote whether line ij is strengthened. If it is strengthened, then Denote the maximum budget for line reinforcement.

[0172] As Figure 4 shown, a decision - making system for strengthening distribution network lines considering typhoon disasters includes a line fault probability generation module, a fault state acquisition module, a load data generation module, an electric vehicle initial module, a restoration model establishment module, a repair time calculation module, a Copeland ranking module, and a decision - making module:

[0173] The line fault probability generation module is used to generate the fault probabilities of each line after a typhoon disaster according to the distribution network pole fault probability generation method;

[0174] The fault state acquisition module is used to obtain the fault states of each line in the distribution network after a typhoon disaster by using non - sequential Monte Carlo simulation based on the fault probabilities of each line after a typhoon disaster;

[0175] The load data generation module is used to generate the non - electric vehicle user load data after a typhoon disaster according to the distribution network topology data and the normal load data of non - electric vehicle users;

[0176] The initial module of the electric vehicle is used to generate the initial state of the electric vehicle after the typhoon disaster according to the parameters of the electric vehicle;

[0177] The restoration model establishment module is used to establish a post-disaster flexible resource collaborative rapid restoration model according to the fault states of each line in the distribution network after the typhoon disaster, the non-electric vehicle user load data after the typhoon disaster, the initial state of the electric vehicle after the typhoon disaster, the objective function of the post-disaster restoration process, and the constraint conditions of the distribution network fault scenario;

[0178] The repair time calculation module is used to simulate the restoration process after the typhoon disaster according to the post-disaster flexible resource collaborative rapid restoration model to obtain the repair time of each line;

[0179] The Copeland ranking module is used to rank according to the repair time of each line using the Copeland ranking method to obtain the line importance ranking;

[0180] The decision-making module is used to formulate the anti-typhoon reinforcement decision of the lines in the distribution network according to the line importance ranking.

[0181] The content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

Claims

1. A decision-making method for strengthening distribution network lines under typhoon disasters, characterized in that: It includes the following steps; Generate the fault probabilities of each line after the typhoon disaster according to the fault probability generation method of distribution network poles; Based on the fault probabilities of each line after the typhoon disaster, use non-sequential Monte Carlo simulation to obtain the fault states of each line in the distribution network after the typhoon disaster; Generate the non-electric vehicle user load data after the typhoon disaster according to the distribution network topology data and the normal load data of non-electric vehicle users; Generate the initial state of electric vehicles after the typhoon disaster according to the electric vehicle parameters; Based on the fault states of each line in the distribution network after the typhoon disaster, the non-electric vehicle user load data after the typhoon disaster, the initial state of electric vehicles after the typhoon disaster, the objective function of the post-disaster recovery process, and the constraint conditions of the distribution network fault scenario, establish a post-disaster flexible resource collaborative rapid recovery model; Simulate the recovery process after the typhoon disaster according to the post-disaster flexible resource collaborative rapid recovery model to obtain the repair times of each line; Sort according to the repair times of each line using the Copeland ranking method to obtain the line importance ranking; Formulate the line typhoon prevention and reinforcement decision of the distribution network according to the line importance ranking.

2. The decision-making method for reinforcing distribution network lines under typhoon disasters according to claim 1, characterized in that The specific method for the fault probability generation method of distribution network poles to generate the fault probabilities of each line is as follows: The fault probability caused by the impact of the typhoon on the poles: where a' is a pole on line ij; w(t) represents the wind speed at time t is the failure probability of pole a'; μ is the resistance of the pole, ξ is the logarithmic standard deviation of the typhoon intensity measurement; p ij is the failure probability of line ij; m is the number of poles on line ij.

3. The decision-making method for reinforcing distribution network lines under typhoon disasters according to claim 1, characterized in that The specific method for using non-sequential Monte Carlo simulation to obtain the fault states of each line in the distribution network after the typhoon disaster based on the fault probabilities of each line after the typhoon disaster and the fault probabilities of each line after the typhoon disaster is as follows: Extract the state of each line of the system by non-temporal Monte Carlo method, and extract a random number rand i , which is uniformly distributed in the [0, 1] space, to obtain the operating state s of line ij at the beginning of the repair process after the typhoon disaster i (0): p ij is the fault probability of line ij.

4. The decision-making method for strengthening the distribution network lines under typhoon disasters according to claim 1, wherein The specific method for generating the non-electric vehicle user load data after the typhoon disaster according to the distribution network topology data and the normal load data of non-electric vehicle users is as follows: Suppose the load of power system node i is the product of a first random multiplier and a second random multiplier with the hourly normalized load curve: where \(N\) represents the set of power system nodes; and \(\overline{P}_{i}\) represents the average of the daily peak active power of power system node \(i\) in a year, \(\overline{Q}_{i}\) represents the average of the daily peak reactive power of power system node \(i\) in a year; where Τ is the set of time; P i L (t) is the active power load of the non - electric vehicle load at node i of the power system at time t, is the reactive power load of the non - electric vehicle load at node i of the power system at time t; M p (t) is the normalized reactive power load curve varying with time, M q (t) is the normalized reactive power load curve varying with time; assume that the time shift of the load curve is uniformly distributed.

5. The decision-making method for reinforcing distribution network lines under typhoon disasters according to claim 1, characterized in that The specific method for generating the initial state of electric vehicles after the typhoon disaster according to the electric vehicle parameters is as follows: The electric vehicle parameters include the initial position of the electric vehicle, the time when the electric vehicle enters the charging station, the daily driving mileage of the electric vehicle, and the battery capacity of the electric vehicle; The initial positions of electric vehicles are uniformly distributed according to the distance to the nearest V2G vehicle grid-connected charging station O t ; The time when the electric vehicle enters the charging station follows a normal distribution, and its probability distribution is: where t ch is the time when the electric vehicle enters the charging station; μ ch is the mean value of the time when entering the charging station; σ ch is the standard deviation of the time when entering the charging station; The daily driving mileage of the electric vehicle follows a lognormal distribution, and its probability distribution is: where s is the daily mileage; μ s is the mean of the daily mileage; σ s is the standard deviation of the daily mileage; The battery capacities of different types of electric vehicles follow a normal distribution, and its probability distribution is: where C p is the battery capacity of the electric vehicle; μ c is the mean value of the battery capacity; σ c is the standard deviation of the battery capacity; The total load of power system node i is the sum of the load of electric vehicle users and the load of non - electric vehicle users, that is P i L,E (t) is the electric vehicle user load of the charging station at power system node i.

6. The decision-making method for strengthening the distribution network lines under typhoon disasters according to claim 1, characterized in that The specific method for establishing a post-disaster flexible resource collaborative rapid recovery model based on the fault states of each line in the distribution network, the non-electric vehicle user load data, the initial state of electric vehicles, the objective function of the post-disaster recovery process, and the constraint conditions of the distribution network fault scenario is as follows: The goal of establishing a post-disaster flexible resource collaborative rapid recovery model is to minimize the load shedding amount during the entire post-disaster recovery process. The objective function of the post-disaster recovery process is: Among them represents the weight of power system node i; P shed,i,t represents the active power load shed of power system node i at time t; where Δt is the time step, and it is assumed that Δt is 15 minutes The constraint conditions of the distribution network fault scenario include line working state constraints, distribution network power flow constraints, radiation state topology constraints, fault impact transfer constraints under distribution network N-k faults, mobile energy storage constraints, demand response constraints, electric vehicle constraints, and repair personnel path constraints. This invention needs to satisfy all constraint conditions simultaneously; The line working state constraint is: where s ij (t) is a 0-1 variable, and s ij (t) = 0 when the line ij fails; s ij (t) = 1 when it works properly; (i, j) ∈ E' is the set of faulty lines; (i, j) ∈ E is the set of lines; The distribution network power flow constraint is: The linearized DistFlow model is used to construct the operation constraints; the power balance constraints of each power system node are: where $N$ represents the set of power system nodes; the node in the inflow direction of power system node $i$ is the parent node of power system node $i$, and $\lambda(i)$ is defined as the set of parent nodes of power system node $i$; the node in the outflow direction of power system node $i$ is the child node of power system node $i$, and $\delta(i)$ is defined as the set of child nodes of power system node $i$; $P$ ij,t is the active power flowing through line $(i, j)$ at time $t$; $Q$ ij,t represents the reactive power flowing through line $(i, j)$ at time $t$; $P$ L,i,t represents the active load of power system node $i$ at time $t$; $Q$ L,i,t represents the reactive load of power system node $i$ at time $t$; $P$ DG,i,t represents the active power output of DG at power system node $i$ at time $t$; $Q$ DG,i,t represents the reactive power output of DG at power system node $i$ at time $t$; $P$ shed,i,t represents the active load shedding of power system node $i$ at time $t$; $Q$ shed,i,t represents the reactive load shedding of power system node $i$ at time $t$; If the branch is closed, the voltage difference of the branch is subject to the power flow constraint. If the branch is open, the voltage difference is arbitrary, the branch flow must be zero, and the voltage relationship between adjacent nodes is constrained as follows: Among them, U i,t represents the voltage of power system node i at time t, and U j,t represents the voltage of power system node j at time t; U0 represents the rated voltage value; r ij and x ij represent the resistance and reactance of line (i, j); c ij,t is a 0-1 variable of the state of line (i, j) at time t, where 0 means disconnected and 1 means closed; M is a first constant, and the value range of M is [10 3 , +∞]; The capacity limit constraint for each line is: Among them represents the maximum allowable power flowing through line (i, j); The maximum and minimum voltage constraints of each power system node are: where $N\{DG\}$ represents the set of nodes without distributed generation installed; and are the maximum and minimum voltage limits at power system node $i$; The output limit constraints of distributed power generation are: where {DG} represents the set of nodes where distributed generation is installed; and is the minimum active power limit of DG at power system node i; is the maximum active power limit of DG at power system node i; is the minimum active power limit of DG at power system node i; is the maximum active power limit of DG at power system node i; is a 0-1 variable indicating whether power system node i is affected by a fault. When power system node i is affected by a fault takes 1, otherwise takes 0; The demand limit constraint of the controllable load is: The radiation state topological constraint is: where N node represents the number of nodes; N sub represents the number of substations; γ i,t is a 0-1 variable, indicating whether a node is a source node; Variable γ i,t It is only allowed to take the value of 1 when the line connected to power system node i is disconnected. The relevant constraint is as follows: Each subgraph is a connected graph, and the relevant constraints are: where {Sub} represents the set of substation nodes; N\{Sub} represents the set of nodes excluding substation nodes, and N node represents the number of nodes; N sub represents the number of substations; γ i,t represents a 0-1 variable indicating whether a node is a source node; V ij,t represents the virtual power of line (i,j) at time t; W i,t represents the output of the virtual power source at power system node i, which is an unrestricted real number. In the virtual network, substation nodes and nodes with γ i,t = 1 are regarded as source nodes, and an unrestricted virtual power source output power is set. For other nodes, a virtual load with a load value of 1 is set; The fault impact transmission constraint under the fault of distribution network Nk is: where ε is the second constant, and the value range of ε is [0, 0.01]; y ij1,t is a binary variable representing the switch state on the i side of line (i, j) at time period t, y ij2t is a binary variable representing the switch state on the j side of line (i, j) at time period t, taking 1 if no switch is configured or the switch is in the closed state, otherwise taking 0; is a binary parameter, taking 1 if RCS is configured on the i side (κ = 1) or j side (κ = 2) of line (i, j), otherwise taking 0; is a 0-1 variable indicating whether node i is affected by a fault. When the node is affected by a fault takes 1, otherwise takes 0; The mobile energy storage constraints are: where {MESS} is the mobile energy storage set; {dep} is the set of warehouses from which the mobile energy storage departs; N MESS is the total number of mobile energy storages; d i,e,t is whether the mobile energy storage e is at the power system node i at time t, which is a 0-1 variable; is the routing time for the mobile energy storage to move from the power system node i to the power system node j; Among them is the upper limit of the state of charge of the mobile energy storage e is the lower limit of the state of charge of the mobile energy storage e; E e,t is the state of charge of the mobile energy storage e at time t; σ e is the self-discharge rate of the mobile energy storage e, η e is the charge-discharge efficiency of the mobile energy storage e; is the active power output of the mobile energy storage e at the power system node i at time t is the reactive power output of the mobile energy storage e at the power system node i at time t; represents the total output of the mobile energy storage at the power system node i at time t; is a 0-1 variable representing the working state of the mobile energy storage e at node i at time t; represents the minimum active power of the mobile energy storage e represents the minimum reactive power of the mobile energy storage e; represents the maximum active power of the mobile energy storage e represents the maximum reactive power of the mobile energy storage e; The demand response constraints are: Among them is the demand response load participated by power system node i at time t; P L,i,t represents the active power load of power system node i at time t; P i max represents the maximum limit value of the active power load of power system node i; is the load of power system node i after participating in demand response at time t; is a 0-1 variable indicating whether demand response device is configured for power system node i; α i,t is the participation ratio of demand response load at time t; P i LD,max is the maximum allowable load reduction of power system node i; The electric vehicle constraints are: The trip chain theory is used to characterize the operating characteristics of EVs; the trip chain set expression is: L = {B0, B f , W 0f , L 0f , T0, T f , T 0f , T p} (38) where L is the set of trip chains; the set elements are the trip chain start point B0, the trip chain end point B f , the trip path W 0f , the corresponding driving distance L of the path 0f , the departure time T0, the arrival time T f , the driving time T 0f and the stopping time T p ; The time constraints for electric vehicles to arrive at the vehicle V2G station are as follows: where t 0,n represents the departure time of the nth journey segment; represents the arrival time of the nth journey segment; represents the driving time of the nth journey segment; is the stop time of the nth journey segment; Constraints on electric vehicle travel willingness: where is the number of EVs participating in the scheduling response; β is the proportion of EVs participating in the restoration; N EV is the total number of EVs; Electric vehicle real-time status constraints: Among them is the set of real-time information of each EV at time t, including the remaining power of the current EV driving state w t and the number O of the V2G station closest to the location t ; Electric vehicle cluster battery state of charge constraints: Among them is the state of charge of electric vehicle k at time t; is the charging and discharging efficiency of electric vehicle k; is the output of electric vehicle k at power system node i; is a 0-1 variable indicating whether electric vehicle k is at power system node i; {EV} represents the set of electric vehicles; Δt is the time step; V2G reverse charging output constraints: Among which P i V2G,max and P i V2G,min are respectively the maximum and minimum outputs of the V2G station located at the power system node i; is the output of the V2G station located at the power system node i at time period t; is the maximum number of vehicles that the V2G station located at the power system node i can accommodate; V G represents the set of V2G stations; The maintenance personnel path constraint is: Among them, Crew represents the set of maintenance teams; {dep} represents the set of departure positions of the maintenance teams; represents the moment when maintenance team c arrives at line m; The line can only be repaired after the team arrives, so the relevant constraints are: where y m,c indicates whether the repair team c has reached line m; f m,t indicates whether line m has been repaired at time t, equal to 1 if so, otherwise equal to 0; The time when the maintenance team arrives at the line and f m,t The relationship is as follows: Among them represents the time required for maintenance team c to repair line m; For coupling f m,t with the line state s m (t) is constrained by: All maintenance teams can only go to the faulty line once and can only leave once. The relevant constraints are as follows: Crew means the maintenance team assembly; {dep} the maintenance team departure position assembly; $E' \cup \{dep\} \setminus \{n\}$ represents the set of damaged lines and depots excluding line $n$; $x$ m,n,c is a 0-1 variable indicating whether repair team $c$ moves from line $m$ to line $n$. It is equal to 1 if so, and 0 otherwise; represents the time consumed for repair team $c$ to move from line $m$ to line $n$; The Big M method is used to make the route of the maintenance personnel continuous. The relevant constraints are as follows: Among them represents the time taken for repair team c to move from line m to line n; represents the moment when repair team c arrives at line m; represents the moment when repair team c arrives at line n; represents the time required for repair team c to repair line m; represents the time taken for repair team c to move from line m to line n; y m,c represents whether repair team c has arrived at line m.

7. The decision-making method for reinforcing distribution network lines under typhoon disasters according to claim 1, characterized in that The recovery process after a typhoon disaster is simulated based on the post-disaster flexible resource collaborative rapid recovery model, and the specific method for obtaining the repair time of each line is as follows: The repair time of each line is calculated by the time when the maintenance personnel arrive at the line and the maintenance time of the line: Among which T m represents the completion time of the repair of line m.

8. The decision-making method for reinforcing distribution network lines under typhoon disasters according to claim 1 is characterized in that, The specific method of sorting the line importance by using the Copeland sorting method according to the repair time of each line is as follows: Copeland score S of line m m,k The calculation method is as follows: where q k (m) represents the k-th percentile of the cumulative distribution function of the repair time of line m; S m,n,k represents the Copeland score after the k-th comparison between line m and line n; S m Indicates the Copeland total score of line m; First, compare line m with all other lines Ω times using the Copeland pairwise comparison method to obtain S m,n,k ; Then add the Copeland scores S m,n,k of each pairwise comparison to get the final Copeland score S m ; Sort the line importance according to the Copeland score S of the line m .

9. The decision-making method for reinforcement of distribution network lines under typhoon disasters according to claim 1, wherein The specific method for generating the failure probability of each line in the distribution network is as follows: The lines are sorted according to their importance, and the top-ranked lines are identified. The supporting poles of the lines are regarded as the distribution network poles that need to be reinforced. The number of reinforcements is determined according to the reinforcement budget: Among them represents the cost of all poles for strengthening the support line ij; represents whether the support line ij is strengthened. If it is strengthened, then represents the maximum budget for line strengthening.

10. A decision-making system for strengthening distribution network lines under typhoon disasters, characterized in that, It includes line fault probability generation module, fault status acquisition module, load data generation module, electric vehicle initialization module, recovery model establishment module, repair time calculation module, Copeland sorting module and decision module: The line fault probability generation module is used to generate the fault probability of each line after the typhoon disaster occurs according to the distribution network line fault probability generation method; The fault status acquisition module is used to obtain the fault status of each line in the distribution network after the typhoon disaster occurs by using non-sequential Monte Carlo simulation according to the fault probability of each line after the typhoon disaster occurs; The load data generation module is used to generate the load data of non-electric vehicle users after the typhoon disaster occurs according to the distribution network topology data and the normal load data of non-electric vehicle users; The electric vehicle initialization module is used to generate the initial state of the electric vehicle after the typhoon disaster occurs according to the electric vehicle parameters; The restoration model establishment module is used to establish a post-disaster flexible resource collaborative rapid restoration model according to the fault status of each line in the distribution network after the typhoon disaster, the non-electric vehicle user load data after the typhoon disaster, the initial state of electric vehicles after the typhoon disaster, the objective function of the post-disaster restoration process, and the constraint conditions of the distribution network fault scenario; The repair time calculation module is used to simulate the restoration process after the typhoon disaster according to the post-disaster flexible resource collaborative rapid restoration model to obtain the repair time of each line; The Copeland ranking module is used to rank according to the repair time of each line using the Copeland ranking method, so as to obtain the line importance ranking; The decision-making module is used to formulate the line typhoon prevention and reinforcement decision of the distribution network according to the line importance ranking.

Images