Electric power first-aid repair deployment method under extreme meteorological disasters

Abstract

The invention discloses an electric power first-aid repair deployment method under extreme meteorological disasters. The method comprises the following steps: firstly, determining the urgency degree of an affected point; secondly, constructing an electric power first-aid repair deployment model; solving the power repair deployment model on the basis of an improved NSGA-II algorithm; and finally, performing example simulation on the power repair deployment model. According to the electric power first-aid repair allocation method under the extreme meteorological disaster, the adaptation condition of the ability of the first-aid repair personnel and the fault condition of the disaster site is fully considered, so that the dispatched first-aid repair personnel can deal with the fault problem encountered by the disaster site, and a plurality of sets of electric power first-aid repair allocation schemes can be obtained; a decision maker can select a scheme with the shortest delivery time or a scheme with the best urgency degree or a scheme with the higher demand satisfaction degree or a compromise of the three schemes according to the actual rescue situation.

Description

Technical Field

[0001] The invention relates to the technical field of emergency repair of electric power systems, and in particular to an emergency repair and dispatching method for electric power under extreme meteorological disasters. Background Art

[0002] After an extreme meteorological disaster occurs, a scientific and efficient plan for the deployment of emergency personnel and materials can greatly reduce the losses caused by the disaster to the power system and avoid serious secondary consequences. In recent years, many scholars at home and abroad have achieved many research results in the field of emergency deployment. Among them, distribution planning is the key issue in formulating emergency material deployment plans. Distribution planning includes personnel planning, route planning, and material distribution planning. The current distribution planning modeling mainly takes urgency weighted time, transportation cost or failure loss as the target optimization conditions, and takes storage reserves, number of repair personnel, emergency demand and transportation capacity as constraints, and applies heuristic algorithms to optimize the target.

[0003] First, the models proposed so far lack the evaluation of the skills and capabilities of the repair team in terms of personnel deployment. The generalized evaluation of repair personnel will lead to the mismatch between the repair team and the repair site, thus delaying rescue. Secondly, most of the current models are single-objective optimization, or focus on optimizing scheduling time or reducing transportation costs, and lack consideration of factors such as fairness or satisfaction. Third, the current deployment model focuses on either personnel deployment or material deployment, and lacks a comprehensive deployment model of the two. Therefore, the emergency deployment plan obtained is too absolute and does not have a strong objective guiding significance for the actual situation. Summary of the invention

[0004] The purpose of the present invention is to provide a method for emergency repair and deployment of electric power under extreme meteorological disasters, which takes into account both personnel deployment and material deployment, and is more comprehensive in emergency decision-making; fully considers the adaptation of the emergency repair personnel's capabilities to the fault conditions at the disaster site, so that the dispatched emergency repair personnel can handle the fault problems encountered at the disaster site.

[0005] To achieve the above object, the present invention provides a method for emergency repair and deployment of electric power under extreme weather disasters, comprising the following steps:

[0006] S1. Determination of the urgency of the disaster site;

[0007] S2. Construct a power repair and dispatch model;

[0008] S3, solving the power repair dispatch model based on the improved NSGA-II algorithm;

[0009] S4. Carry out example simulation on the power repair and dispatch model.

[0010] Preferably, in S1, the urgency is determined based on three priority indicators: the importance of power equipment S, the importance of power users L, and the difficulty of power repair D;

[0011] Among them, power equipment refers to power transmission and transformation equipment, power users refer to party and government agencies, hospitals, radio stations and transportation hubs, and the difficulty of power emergency repairs includes no need for emergency repairs, need for emergency repairs and paralyzed.

[0012] Preferably, in S1, determining the urgency comprises the following steps:

[0013] S11. Establish a scoring table;

[0014] A scoring table is established based on the importance of the facility S, the importance of the user L and the difficulty of power repair D, and the scoring value is between 0 and 10;

[0015] S12, calculating the urgency;

[0016] Based on S11, the urgency u of disaster site j j It is expressed as:

[0017] u j =aS j +bL j +cD j ;

[0018] According to the requirements of the emergency plan, weights are determined for the three priority indicators of power equipment importance S, power user importance L and power repair difficulty D. The corresponding weights are a = 0.75, b = 0.15, c = 0.1, and the evaluation model is expressed as:

[0019]

[0020] In the formula, when D=0, it means that the equipment has been paralyzed or the road is blocked and emergency repair is impossible; when D=10, it means that the equipment is basically not damaged and no emergency repair is needed.

[0021] Preferably, in S2, the power emergency repair deployment model includes a power emergency repair personnel deployment model and a power emergency repair material deployment model;

[0022] The construction process of the power repair personnel deployment model is as follows:

[0023] According to the fault type of the disaster site, the corresponding repair team is dispatched. The fitness reflects the degree of fit between the repair team's capabilities and the fault type of the disaster site. The steps for determining the fitness are as follows:

[0024] S2a1, determine the repair team's capability vector and the disaster-affected point's fault vector;

[0025] The capability vector of the repair team i is:

[0026] r i =[r i1 ,r i2 ,.....,r iK ];

[0027] Among them, r ik It represents the number of people who can handle the kth type of fault, K is the number of faults;

[0028] Failure demand vector of disaster site j:

[0029] d j =[d j1 ,d j2 ,.....,d jK ];

[0030] Among them, d jk represents the number of people required to recover from the kth fault, where K is the number of faults;

[0031] S2a2, determination of fitness function;

[0032] The fitness function is as follows:

[0033]

[0034] in:

[0035]

[0036] Considering shortening the time to reach the repair point, the objective function is as follows:

[0037]

[0038] Where Z is the sum of the urgency weighted fitness, x ij is the decision variable, x ij =1 means that the i-th repair team is dispatched to the j-th disaster site, x ij =0 means no dispatch, T1 means the total time spent on personnel delivery, t ij The time it takes to dispatch from the i-th disaster site to the j-th disaster site;

[0039] The constraints of the objective function are as follows:

[0040] (1) The number of personnel dispatched to handle fault k at disaster site j should not be less than the required number:

[0041]

[0042] Where: p ijk It represents the number of people sent by the i-th team to the j-th disaster site to handle the k-th type of fault;

[0043] (2) The number of personnel dispatched to handle fault k should not be greater than the total number of personnel who can handle fault k:

[0044]

[0045] (3) Every disaster site must have personnel to repair it:

[0046]

[0047] (4) The value of the dispatched decision variable can only be 0 or 1:

[0048] x ij ={0,1}.

[0049] Preferably, in S2, the construction process of the power repair material allocation model is:

[0050] Taking into account the urgency, a power emergency material allocation model is constructed with the goal of maximizing the fairness of emergency material distribution and minimizing the total delivery time. The objective function is as follows:

[0051]

[0052] In the formula, s represents the weighted undersatisfaction sum of material urgency, T2 represents the total transportation time of material distribution, and b j represents the material demand of disaster site j, a j represents the delivery volume of the distribution center, x ij is the decision variable, x ij =1 means that in the transportation route, after visiting the disaster-stricken point i, the first disaster-stricken point j should be visited. ij represents the time taken to travel from disaster site i to disaster site j, and t represents the unloading time of a unit of materials;

[0053] The constraints of the objective function are as follows:

[0054] (1) The transport volume shall not exceed the demand but shall not be less than 50% of the demand:

[0055] 0.5b j ≤a j ≤ b j ;

[0056] (2) All disaster-affected points are passed through once and only once:

[0057]

[0058] Where Q is the set of disaster-affected points.

[0059] Preferably, in S3, solving the power emergency repair dispatch model based on the improved NSGA-II algorithm is specifically as follows:

[0060] In the power repair personnel deployment model, the chromosome will be designed as an M×N×P matrix, where M is the number of fault types, N is the number of disaster-affected points, and P is the number of repair teams;

[0061] In the power repair material allocation model, the chromosome will be designed as a 2×N matrix, where N is the number of disaster-affected points. The first row of the 2×N matrix is ​​the transportation route consisting of the disaster-affected point numbers; the second row is the number of materials dropped at the disaster-affected point corresponding to the first row;

[0062] Each chromosome is a deployment plan. The population will select individuals for crossover and mutation operations based on the non-dominated level and crowding degree. After several rounds of iterations, the Pareto frontier solution set is obtained, and the decision maker selects the plan to execute based on the actual situation.

[0063] Preferably, in S3, solving the power emergency repair dispatch model based on the improved NSGA-II algorithm includes the following steps:

[0064] S31. Population initialization:

[0065] Randomly generate the initial population, initialize the population size and the number of iterations G;

[0066] S32, fitness evaluation:

[0067] The model objective function is the fitness function, which evaluates the performance of individuals on all objectives. When an individual cannot meet all constraints, its fitness will be penalized. The penalty function is as follows:

[0068] f = fp·penalty;

[0069] In the formula, f is the fitness function value, p is the penalty factor, and penalty is the degree of violation of the constraint condition;

[0070] S33, non-dominated sorting:

[0071] Perform non-dominated sorting on the population to generate multiple levels F1, F2, .., each level contains a set of non-dominated solutions; for each individual, record its dominance number and the individuals it dominates;

[0072] S34. Calculate the congestion distance:

[0073] For each level, the crowding distance of each solution is calculated to evaluate the distribution density of the solution in the target space;

[0074] S35, selection, crossover and mutation:

[0075] The tournament selection method is adopted, and the non-dominated level is given priority in selection, and the crowding distance is given second priority. The selected parents are subjected to crossover and mutation operations to generate offspring individuals.

[0076] S36. Merge populations:

[0077] Merge the parent and child populations, and recalculate the non-dominated sorting and crowding distance;

[0078] S37. Select a new generation of population:

[0079] Select the best individuals according to non-dominated sorting and crowding distance to form a new population;

[0080] S38. Repeat S32 to S37. When the number of iterations reaches the maximum number of generations, stop the algorithm and output the Pareto frontier solution set.

[0081] Therefore, the present invention adopts the above-mentioned power repair and deployment method under extreme meteorological disasters, and the beneficial effects are as follows:

[0082] (1) The power emergency repair and deployment method provided by the present invention takes into account both personnel deployment and material deployment, and is more comprehensive in emergency decision-making.

[0083] (2) The power emergency repair and dispatching method provided by the present invention fully considers the adaptation between the emergency repair personnel's capabilities and the fault conditions at the disaster site, so that the dispatched emergency repair personnel can handle the fault problems encountered at the disaster site.

[0084] (3) The power emergency repair and dispatching method provided by the present invention can obtain several sets of power emergency repair and dispatching plans. Decision makers can choose the plan with the shortest delivery time, the plan with the best urgency, the plan with higher demand satisfaction, or a compromise among the three according to the actual rescue situation.

[0085] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 This is a flow chart of the NSGA-II algorithm of an embodiment of a method for emergency repair and dispatching of electric power under extreme meteorological disasters of the present invention;

[0087] Figure 2 It is a Pareto frontier image of personnel deployment in an embodiment of a method for power repair and deployment under extreme meteorological disasters of the present invention;

[0088] Figure 3 It is a material allocation Pareto front image of an embodiment of a method for emergency repair and allocation of electric power under extreme meteorological disasters of the present invention. DETAILED DESCRIPTION

[0089] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0090] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0091] like Figure 1 As shown, a method for emergency repair and dispatching of electric power under extreme weather disasters comprises the following steps:

[0092] S1. Determine the urgency of the disaster site.

[0093] After an extreme weather disaster occurs, power companies should invest limited emergency repair forces and rescue materials as much as possible in places with high urgency (such as substations, hospitals, etc.) to avoid greater losses and derivative accidents. According to the relevant provisions of the "State Grid Corporation of China's Emergency Response Rules for Power Emergencies", the urgency should be determined based on the three priority indicators of power equipment importance S, power user importance L and power emergency repair difficulty D. Among them, important power equipment refers to power transmission and transformation equipment, such as substations, important power users refer to party and government agencies, hospitals, radio stations and transportation hubs, and the difficulty of power emergency repair includes no need for emergency repair, need for emergency repair and paralyzed.

[0094] Urgency determination involves the following steps:

[0095] S11. Create a scoring table.

[0096] The power company organizes experts and security department personnel to establish a scoring table based on the importance of the facility S, the importance of the user L and the difficulty of power repair D. As shown in Table 1, the scoring value is between (0-10):

[0097] Table 1 Example of scoring table

[0098]

[0099]

[0100] S12. Calculate the urgency.

[0101] Based on S11, the urgency u of disaster site j j It can be expressed as:

[0102] u j =aS j +bL j +cD j ;

[0103] By consulting relevant national policy documents and the emergency plan requirements of power companies in various provinces, cities and districts, the following conclusions can be drawn:

[0104] (1) The level of power transmission and transformation equipment and facilities should be absolutely more important than the importance of users and the difficulty of emergency repairs;

[0105] (2) The importance of users should be significantly more important than the difficulty of repair;

[0106] (3) The difficulty of emergency repair should be the last factor to be considered.

[0107] According to the requirements of the emergency plan, weights can be determined for the three priority indicators: power equipment importance S, power user importance L, and power repair difficulty D. The corresponding weights are a = 0.75, b = 0.15, and c = 0.1, respectively. The evaluation model is expressed as:

[0108]

[0109] In the formula, when D=0, it means that the equipment has been paralyzed or the road is blocked and emergency repair is impossible; when D=10, it means that the equipment is basically not damaged and no emergency repair is needed.

[0110] S2. Construct a power emergency repair and dispatch model.

[0111] The power repair and dispatch model includes two sub-models: the power repair personnel dispatch model and the power repair material dispatch model.

[0112] The construction process of the power repair personnel deployment model is as follows:

[0113] After an extreme weather disaster, common fault types in the power system include cable faults, line faults, and secondary system faults. Therefore, decision makers should dispatch corresponding repair teams according to the fault types at the disaster site, and use fitness to reflect the degree of fit between the repair team's capabilities and the fault types at the disaster site. A high degree of mutual fitness indicates that the repair team can better handle fault problems at the disaster site.

[0114] Among them, the steps for determining the fitness function are as follows:

[0115] S2a1, determine the repair team's capability vector and the disaster-affected point's fault vector;

[0116] The capability vector of the repair team i is:

[0117] r i =[r i1 ,r i2 ,.....,r iK ];

[0118] Among them, r ik It represents the number of people who can handle the kth type of fault, K is the number of faults;

[0119] Failure demand vector of disaster site j:

[0120] d j =[d j1 ,d j2 ,.....,d jK ];

[0121] Among them, d jk represents the number of people required to recover from the kth fault, where K is the number of faults;

[0122] S2a2, determination of fitness function;

[0123] The fitness function is as follows:

[0124]

[0125] in:

[0126]

[0127] After an extreme weather disaster occurs, it is necessary not only to comprehensively consider the urgency and adaptability, but also to shorten the time to reach the repair point as much as possible. Based on this, the objective function is as follows:

[0128]

[0129] Where Z is the sum of the urgency weighted fitness, x ij is the decision variable, x ij =1 means that the i-th repair team is dispatched to the j-th disaster site, x ij =0 means no dispatch, T1 means the total time spent on personnel delivery, t ij The time it takes to dispatch from the i-th disaster site to the j-th disaster site;

[0130] The constraints of the objective function are as follows:

[0131] (1) The number of personnel dispatched to handle fault k at disaster site j should not be less than the required number:

[0132]

[0133] Where: p ijk It represents the number of people sent by the i-th team to the j-th disaster site to handle the k-th type of fault;

[0134] (2) The number of personnel dispatched to handle fault k should not be greater than the total number of personnel who can handle fault k:

[0135]

[0136] (3) Every disaster site must have personnel to repair it:

[0137]

[0138] (4) The value of the dispatched decision variable can only be 0 or 1:

[0139] x ij ={0,1}.

[0140] The construction process of the power repair material allocation model is as follows:

[0141] After an extreme meteorological disaster, the distribution center has limited emergency supplies and cannot meet the material needs of all disaster-stricken sites. Therefore, the design of the power repair material allocation model should take into account time, fairness and urgency. Based on this, the present invention constructs a power emergency material allocation model with the goal of maximizing the fairness of emergency material distribution and minimizing the total distribution time on the basis of considering urgency. The objective function is as follows:

[0142]

[0143] In the formula, s represents the weighted undersatisfaction sum of material urgency, T2 represents the total transportation time of material distribution, and b j represents the material demand of disaster site j, a j represents the delivery volume of the distribution center, x ij is the decision variable, x ij =1 means that in the transportation route, after visiting the disaster-stricken point i, the first disaster-stricken point j should be visited. ij It represents the time taken to travel from disaster-affected point i to disaster-affected point j, and t represents the unloading time of unit materials.

[0144] The constraints of the objective function are as follows:

[0145] (1) The transport volume shall not exceed the demand but shall not be less than 50% of the demand:

[0146] 0.5b j ≤a j ≤ b j ;

[0147] (2) All disaster-affected points are passed through once and only once:

[0148]

[0149] Where Q is the set of disaster-affected points.

[0150] S3. Solve the power repair dispatch model based on the improved NSGA-II algorithm, specifically:

[0151] In the power repair personnel deployment model, the chromosome will be designed as an M×N×P matrix, where M is the number of fault types, N is the number of disaster-affected points, and P is the number of repair teams.

[0152] In the power repair material allocation model, the chromosome will be designed as a 2×N matrix, where N is the number of disaster-stricken points. The first row of the 2×N matrix is ​​the transportation route consisting of the disaster-stricken point numbers; the second row is the number of materials delivered to the disaster-stricken point corresponding to the first row.

[0153] Each chromosome is a deployment plan. The population will select individuals for crossover and mutation operations based on the non-dominated level and crowding degree. After several rounds of iterations, the Pareto frontier solution set is obtained, and the decision maker selects the plan to execute based on the actual situation.

[0154] Solve the power emergency repair dispatch model based on the improved NSGA-II algorithm.

[0155] The NSGA-II algorithm is good at handling multi-objective optimization problems and can effectively balance multiple objectives to find the Pareto optimal solution set. The NSGA-II algorithm is also highly flexible and can adapt to different types of constraints. The model solution includes the following steps:

[0156] S31. Population initialization:

[0157] Randomly generate the initial population, initialize the population size and the number of iterations G, and determine whether the constraints are met. If not, regenerate the individuals.

[0158] S32, fitness evaluation:

[0159] The model objective function is the fitness function, which evaluates the performance of individuals on all objectives. Since the traditional NSGA-II algorithm will directly eliminate individuals that do not meet the constraints, resulting in a smaller population size and slower convergence speed. Therefore, the present invention will introduce a penalty mechanism to protect individuals that do not meet the constraints. That is, when an individual cannot meet all the constraints, its fitness will be penalized. The penalty function is as follows:

[0160] f = fp·penalty;

[0161] In the formula, f is the fitness function value, p is the penalty factor, and penalty is the degree of violation of the constraint.

[0162] S33, non-dominated sorting:

[0163] Perform non-dominated sorting on the population to generate multiple levels F1, F2, ... Each level contains a set of non-dominated solutions; for each individual, record its dominance number and the dominated individual.

[0164] S34. Calculate the congestion distance:

[0165] For each level, the crowding distance of each solution is calculated to evaluate the distribution density of the solution in the target space. The crowding distance is calculated as follows:

[0166]

[0167] In the formula, Crowded i Indicates the non-dominated level F i The degree of congestion; i,m Indicates the non-dominated level F i The crowding distance on the mth target; f im next Indicates the non-dominated level F i The next individual fitness value on the mth target; f im now Indicates the non-dominated level F i The current individual fitness value on the mth target; f max,m Indicates the non-dominated level F i The optimal fitness value on the mth target; f min,m Indicates the non-dominated level F i The worst fitness value on the m-th target.

[0168] S35, selection, crossover and mutation:

[0169] The tournament selection method is adopted, and non-dominated ranks are given priority in selection, and crowding distance is given second priority. Crossover (such as uniform crossover and single-point crossover) and mutation (such as random gene change) operations are performed on the selected parents to generate offspring individuals.

[0170] S36. Merge populations:

[0171] Merge the parent and child populations, and recalculate the non-dominated sorting and crowding distance.

[0172] S37. Select a new generation of population:

[0173] The optimal individuals are selected according to the non-dominated sorting and crowding distance to form a new population.

[0174] S38. Repeat S32 to S37. When the number of iterations reaches the maximum number of generations, stop the algorithm and output the Pareto frontier solution set.

[0175] The Pareto frontier is used to describe the trade-off between different objectives. A set of solutions is called Pareto efficient (or non-dominated). When a solution is improved on a certain objective, it will inevitably lead to the deterioration of other objectives. In other words, one objective cannot be made better by changing a certain solution without making other objectives worse.

[0176] Specifically, the Pareto front is the set of all Pareto-efficient solutions. In two-dimensional space, the Pareto front is usually a curve that represents all possible solutions that cannot be further improved on one objective without sacrificing another objective.

[0177] S4. Carry out example simulation on the power repair and dispatch model.

[0178] Embodiment 1

[0179] Assume that there are 5 disaster-affected sites, 4 types of faults, 5 emergency repair teams, a material distribution center, and a material reserve of 80. There are enough rescue personnel; the material reserves are insufficient to meet all the needs of the disaster-affected sites. This embodiment uses Pycharm Community Edition 2022.1 for programming and solving. The experimental hardware environment is 13th Gen Intel (R) Core (TM) i7-13650HX 2.60GHz, 24G memory, and Windows 11 operating system.

[0180] Power repair personnel deployment model simulation example

[0181] The rescue team's capabilities are described in Table 2 below:

[0182] Table 2 Rescue team capabilities

[0183]

[0184] The fault repair requirements for each disaster site are shown in Table 3:

[0185] Table 3 Rescue team requirements

[0186]

[0187] Solve and get the Pareto frontier image as Figure 2 As shown by Figure 2 It can be found that it is difficult to optimize both goals simultaneously in personnel deployment decisions. When more adaptability and urgency are met, the transportation time will increase. In actual scenarios, decisions can be made based on multiple factors such as the stage of the disaster and the intensity of the disaster.

[0188] If a compromise solution is chosen, the fitness is 3.997, the transportation time is 101.2, and the transportation solution is shown in Table 4:

[0189] Table 4 Distribution plan

[0190]

[0191]

[0192] Power repair material deployment model simulation

[0193] The total demand at the disaster-affected points is greater than the total reserves of the distribution center. The model optimizes the two objectives of "delivery time" and "weighted sum of material satisfaction and urgency".

[0194] The material needs of the disaster-affected areas are shown in Table 5:

[0195] Table 5 Material demand in disaster-affected areas

[0196] Disaster Area Number One Number Two Number Three Number Four Number Five Demand 30 20 25 15 10

[0197] The Pareto frontier diagram is shown in Figure 3. Figure 3 It can be seen that when the transportation time is shortened to improve efficiency, the "weighted sum of material satisfaction and urgency" will decrease. It is difficult for the model to optimize both goals at the same time. Therefore, decision makers need to fully consider the attributes of the disaster site and rescue arrangements based on the actual situation.

[0198] If you want to achieve a relatively high satisfaction without significantly increasing the time, you can choose a solution with a satisfaction of 75%, and the total time consumption is 1250. The solution is as follows:

[0199] Table 6 Material distribution plan

[0200] Disaster Area Number One Number Two Number Three Number Four Number Five Delivery volume 28 17 13 10 7 Delivery Order 1 3 5 2 4

[0201] Therefore, the present invention adopts the above-mentioned power repair and deployment method under extreme meteorological disasters to obtain several sets of power repair and deployment plans. Decision makers can choose the plan with the shortest delivery time, the plan with the best urgency, the plan with higher demand satisfaction, or a compromise among the three according to the actual rescue situation.

[0202] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for emergency repair and deployment of electric power in extreme weather disasters, characterized in that: The following steps are involved: S1. Determination of the urgency of the disaster site; S2. Construct a power repair and dispatch model; S3, solving the power repair dispatch model based on the improved NSGA-II algorithm; S4. Carry out example simulation on the power repair and dispatch model.

2. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 1, characterized in that: In S1, the urgency is determined based on three priority indicators: the importance of power equipment S, the importance of power users L, and the difficulty of power repair D; Among them, power equipment refers to power transmission and transformation equipment, power users refer to party and government agencies, hospitals, radio stations and transportation hubs, and the difficulty of power emergency repairs includes no need for emergency repairs, need for emergency repairs and paralyzed.

3. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 2, characterized in that: In S1, the urgency determination includes the following steps: S11. Establish a scoring table; A scoring table is established based on the importance of the facility S, the importance of the user L and the difficulty of power repair D, and the scoring value is between 0 and 10; S12, calculating the urgency; Based on S11, the urgency u of disaster site j j It is expressed as: u j =aS j +bL j +cD j ; According to the requirements of the emergency plan, weights are determined for the three priority indicators of power equipment importance S, power user importance L and power repair difficulty D. The corresponding weights are a = 0.75, b = 0.15, c = 0.1, and the evaluation model is expressed as: In the formula, when D=0, it means that the equipment has been paralyzed or the road is blocked and emergency repair is impossible; when D=10, it means that the equipment is basically not damaged and no emergency repair is needed.

4. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 3 is characterized in that: In S2, the power repair deployment model includes the power repair personnel deployment model and the power repair material deployment model; The construction process of the power repair personnel deployment model is as follows: According to the fault type of the disaster site, the corresponding repair team is dispatched. The fitness reflects the degree of fit between the repair team's capabilities and the fault type of the disaster site. The steps for determining the fitness are as follows: S2a1, determine the repair team's capability vector and the disaster-affected point's fault vector; The capability vector of repair team i is: r i ＝[r i1 ,r i2 ,.....,r iK ]； Among them, r ik It represents the number of people who can handle the kth type of fault, K is the number of faults; Failure demand vector of affected point j: d j =[d j1 ,d j2 ,.....,d jK ]; Among them, d jk represents the number of people required to recover from the kth fault, where K is the number of faults; S2a2, determination of fitness function; The fitness function is as follows: in: Considering shortening the time to reach the repair point, the objective function is as follows: Where Z is the sum of the urgency weighted fitness, x ij is the decision variable, x ij =1 means that the i-th repair team is dispatched to the j-th disaster site, x ij =0 means no dispatch, T1 means the total time spent on personnel delivery, t ij The time it takes to dispatch from the i-th disaster site to the j-th disaster site; The constraints of the objective function are as follows: (1) The number of personnel dispatched to handle fault k at disaster site j should not be less than the required number: Where: p ijk It represents the number of people sent by the i-th team to the j-th disaster site to handle the k-th type of fault; (2) The number of personnel dispatched to handle fault k should not be greater than the total number of personnel who can handle fault k: (3) Every disaster site must have personnel to repair it: (4) The value of the dispatched decision variable can only be 0 or 1: x ij ＝{0,1}。 5. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 4, characterized in that: In S2, the construction process of the power repair material allocation model is as follows: Taking into account the urgency, a power emergency material allocation model is constructed with the goal of maximizing the fairness of emergency material distribution and minimizing the total delivery time. The objective function is as follows: In the formula, s represents the weighted undersatisfaction sum of material urgency, T2 represents the total transportation time of material distribution, and b j represents the material demand of disaster site j, a j represents the delivery volume of the distribution center, x ij is the decision variable, x ij =1 means that in the transportation route, after visiting the disaster-affected point i, the first disaster-affected point j should be visited. ij represents the time taken to travel from disaster site i to disaster site j, and t represents the unloading time of a unit of materials; The constraints of the objective function are as follows: (1) The transport volume shall not exceed the demand but shall not be less than 50% of the demand: 0.5b j ≤a j ≤b j ; (2) All disaster-affected points are passed through once and only once: Where Q is the set of disaster-affected points.

6. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 5, characterized in that: In S3, the power repair dispatch model is solved based on the improved NSGA-II algorithm as follows: In the power repair personnel deployment model, the chromosome will be designed as an M×N×P matrix, where M is the number of fault types, N is the number of disaster-affected points, and P is the number of repair teams; In the power repair material allocation model, the chromosome will be designed as a 2×N matrix, where N is the number of disaster-affected points. The first row of the 2×N matrix is ​​the transportation route consisting of the disaster-affected point numbers; the second row is the number of materials dropped at the disaster-affected point corresponding to the first row; Each chromosome is a deployment plan. The population will select individuals for crossover and mutation operations based on the non-dominated level and crowding degree. After several rounds of iterations, the Pareto frontier solution set is obtained, and the decision maker selects the plan to execute based on the actual situation.

7. The method for emergency repair and deployment of electric power in extreme weather disasters according to claim 6, characterized in that: In S3, solving the power repair dispatch model based on the improved NSGA-II algorithm includes the following steps: S31, population initialization: Randomly generate the initial population, initialize the population size and the number of iterations G; S32, fitness evaluation: The model objective function is the fitness function, which evaluates the performance of individuals on all objectives. When an individual cannot meet all constraints, its fitness will be penalized. The penalty function is as follows: f = fp·penalty; In the formula, f is the fitness function value, p is the penalty factor, and penalty is the degree of violation of the constraint condition; S33, non-dominated sorting: Perform non-dominated sorting on the population to generate multiple levels F1, F2, .., each level contains a set of non-dominated solutions; for each individual, record its dominance number and the individuals it dominates; S34. Calculate the congestion distance: For each level, the crowding distance of each solution is calculated to evaluate the distribution density of the solution in the target space; S35, selection, crossover and mutation: The tournament selection method is adopted, which gives priority to non-dominated rank and second priority to crowding distance. The selected parents are crossover and mutation operations are performed to generate offspring individuals. S36. Merge populations: Merge the parent and child populations, and recalculate the non-dominated sorting and crowding distance; S37. Select a new generation of population: Select the best individuals according to non-dominated sorting and crowding distance to form a new population; S38. Repeat S32 to S37. When the number of iterations reaches the maximum number of generations, stop the algorithm and output the Pareto frontier solution set.

Images