CN-121998320-A - Three-network emergency resource optimization method and system integrating unmanned aerial vehicle communication relay

CN121998320ACN 121998320 ACN121998320 ACN 121998320ACN-121998320-A

Abstract

The invention discloses a three-network emergency resource optimization method and system integrating unmanned aerial vehicle communication relay, belonging to the technical field of power system planning and disaster defense; the method comprises the steps of constructing an outer layer optimization model taking a pre-disaster multi-type emergency resource layout scheme as a decision variable, embedding constraint conditions of a communication recovery basic framework by adopting an improved hawk optimization algorithm, carrying out iterative solution on the outer layer optimization model, determining a most robust pre-disaster resource deployment scheme under all scenes, constructing an inner layer model by adopting mixed integer second order cone planning, carrying out refined solution on post-disaster emergency response, optimizing the pre-disaster resource deployment scheme, and constructing a double-layer robust optimization model to realize three-network emergency resource optimization. The invention can promote the collaborative scheduling capability of the multi-type resources in the post-disaster stage of the pre-disaster and the post-disaster, enhance the guarantee capability of the key load and has global searching capability and engineering adaptability.

Inventors

LIU LE
FU JINLIANG
Li Shichuang
KANG XIAONING
CHEN XUMING

Assignees

西安交通大学

Dates

Publication Date: 20260508
Application Date: 20260113

Claims (10)

1. The three-network emergency resource optimization method integrating the unmanned aerial vehicle communication relay is characterized by comprising the following steps of: S1, establishing a communication power supply coupling model based on unmanned aerial vehicle relay driving to obtain a communication recovery basic framework; S2, constructing an outer layer optimization model taking a pre-disaster multi-type emergency resource layout scheme as a decision variable based on a communication recovery basic framework and an uncertainty fault scene set caused by sudden disasters; s3, adopting an improved hawk optimizing algorithm, embedding constraint conditions of a communication recovery basic framework, carrying out iterative solution on an external optimizing model, and determining a pre-disaster resource deployment scheme which is most robust in all scenes; s4, constructing an inner layer model by adopting mixed integer second order cone planning based on a pre-disaster resource deployment scheme, a communication recovery basic frame and a given disaster scene, and carrying out refined solution on post-disaster emergency response to obtain weighted load recovery levels in different scenes, wherein the weighted load recovery levels reversely drive iterative search of S3, so as to optimize the pre-disaster resource deployment scheme; S5, a pre-disaster deployment scheme output by the outer layer optimization model and a post-disaster scheduling result output by the inner layer model are synthesized, a double-layer robust optimization model is built, and full-period resource optimization of pre-disaster optimization pre-deployment and post-disaster collaborative mobilization of three-network multi-type emergency resources is achieved.
2. The method for optimizing three-network emergency resources for fusing unmanned aerial vehicle communication relays according to claim 1, wherein in the step S1, the communication recovery basic framework comprises time domain reachability, link availability, information gating and unmanned aerial vehicle operation related constraints; the communication power supply coupling model is constructed based on a coverage matrix of topology hops and specifically comprises the steps of defining time domain reachability, link availability and information gating, and forming an unmanned aerial vehicle relay communication frame through constraint coupling with unmanned aerial vehicle electric quantity and charging anchor points, wherein the unmanned aerial vehicle relay communication frame can construct an air-ground multi-hop link, and connectivity can be used as a prerequisite constraint of electric power operation.
3. The three-network emergency resource optimization method for fusing unmanned aerial vehicle communication relays according to claim 1, wherein in the step S2, the uncertainty fault scene set is obtained through a monte carlo simulation method, and the monte carlo simulation method is used for generating a multi-type fault combined scene covering node soaking, line damage, road blocking and off-line of a communication base station, and providing scene input for an outer layer optimization model.
4. The three-network emergency resource optimization method for fusing unmanned aerial vehicle communication relays according to claim 1, wherein in the step S3, the improved bald hawk optimization algorithm is used for generating a random coding key and decoding the random coding key into a multi-point site selection, capacity/programming deployment scheme of multi-type emergency resources on candidate warehouses and prepositions.
5. The three-network emergency resource optimization method for fusing unmanned aerial vehicle communication relays according to claim 4, wherein in the step S3, the solving process of the improved bald eagle optimization algorithm comprises a random initialization phase, adding emergency team dispatch constraints, adding UAV theory and related constraints, introducing the influence of storm flood on traffic and communication networks, calling an inner layer MISOCP for solving and evolution phase for each disaster scene, wherein: generating an initial coding representation solution space in an interval, and adding an emergency team deployment constraint; adding dispatch constraints of an emergency team, and describing the departure position, path continuity, access uniqueness, task completion requirements and fault point processing sequence of the emergency team when the emergency team executes the rush repair task after disaster; establishing a communication power supply coupling model driven by UAV relay, and adding UAV theory and related constraint, wherein the communication power supply coupling model comprises mobility and time domain reachability, reachability indexes, coverage superposition and link availability, coverage matrix definition, communication coverage and information gating and unmanned aerial vehicle operation and charging anchor points; Introducing the influence of storm flood on traffic and communication networks and influencing the running speed of emergency rescue vehicles at the depth of accumulated water; the inner layer MISOCP is called for solving aiming at each disaster scene to obtain a response value of the scheme under the scene, and the evaluation results of different scenes are aggregated to form Lu Bangshi fitness of the candidate scheme; And in the evolution stage, executing a selection, search and disturbance strategy of an improved bald eagle optimization algorithm, continuously improving the distribution of solutions through migration, jerk and drift search mechanisms to enable the solutions to converge towards robust optimal deployment, and finally obtaining a pre-disaster resource deployment scheme which is the most robust in all scenes through continuous iteration by an outer layer.
6. The method for optimizing the three-network emergency resource by fusing the unmanned aerial vehicle communication relays according to claim 1, wherein in the step S4, an inner layer builds a MISOCP model aiming at maximizing the weighted load recovery, and the model is used for simulating the post-disaster rush repair and power supply recovery process in the disaster.
7. The three-network emergency resource optimization method for fusing unmanned aerial vehicle communication relays according to claim 1 is characterized in that in the step S2, inner layer model decision variables comprise site selection, start-stop scheduling and temporary supply time sequences of MES and DG, dispatch paths, arrival time and operation start-stop time of maintenance teams and drainage teams, voltage states of distribution networks, load flow balance, branch capacity and line operation, cross-resource cooperative logic of drainage, overhaul, temporary supply and return networks, single-point mutual exclusion operation relation of DG/MES at the same node, and communication gating of unmanned aerial vehicle relay communication.
8. The three-network emergency resource optimization method for fusing the unmanned aerial vehicle communication relay according to claim 1, wherein in the S4, an inner layer model constraint system comprises flow conservation, unique access constraint and self-loop constraint in team scheduling, arrival, start-up and completion relations in operation time sequence, balance of water discharge and operation duration, and flow conservation and second-order cone capacity limitation of a power distribution network; and finally returning the weighted load recovery level of the inner layer model in the scene for the adaptability evaluation of the outer layer.
9. The three-network emergency resource optimization method for fusing unmanned aerial vehicle communication relays according to claim 1, wherein in S5, the objective function of the two-layer robust optimization model is as follows: ; Wherein, the For the pre-disaster resource layout scheme, The level is restored for the weighted load.
10. The three-network emergency resource optimization system for fusing the unmanned aerial vehicle communication relay is used for realizing the three-network emergency resource optimization method for fusing the unmanned aerial vehicle communication relay according to any one of claims 1-9, and is characterized by comprising a communication recovery module, an outer layer optimization module, a model solving module, an inner layer optimization module and a double-layer robust optimization module, wherein: The communication recovery module is used for establishing a communication power supply coupling model based on the unmanned aerial vehicle relay drive to obtain a communication recovery basic frame; The outer layer optimization module is used for constructing an outer layer optimization model taking a pre-disaster multi-type emergency resource layout scheme as a decision variable based on the communication recovery basic framework and an uncertainty fault scene set caused by the sudden disaster; The model solving module is used for adopting an improved balying optimization algorithm, embedding constraint conditions of a communication recovery basic framework, carrying out iterative solving on an external layer optimization model, and determining a pre-disaster resource deployment scheme which is most robust in all scenes; The inner layer optimization module is used for constructing an inner layer model based on a pre-disaster resource deployment scheme, a communication recovery basic frame and a given disaster scene by adopting mixed integer second order cone planning, carrying out refined solution on post-disaster emergency response to obtain weighted load recovery levels in different scenes, carrying out back-driving iterative search on the weighted load recovery levels, and optimizing the pre-disaster resource deployment scheme; The double-layer robust optimization module is used for integrating a pre-disaster deployment scheme output by the outer layer optimization model and a post-disaster scheduling result output by the inner layer model, constructing a double-layer robust optimization model and realizing full-period resource optimization of pre-disaster optimization pre-deployment and post-disaster collaborative mobilization of three-network multi-type emergency resources.

Description

Three-network emergency resource optimization method and system integrating unmanned aerial vehicle communication relay Technical Field The invention belongs to the technical field of power system planning and disaster prevention, and particularly relates to a three-network emergency resource optimization method and system integrating unmanned aerial vehicle communication relay. Background With the frequent occurrence of extreme natural disasters, sudden events such as floods, earthquakes, landslides and the like cause remarkable multi-network coupling type damage to power distribution networks, power communication networks and traffic transportation networks. The problems of node soaking, line damage, road blocking, off-line of a communication base station and the like caused by disasters have strong synchronicity and relevance, so that the traditional post-disaster rush-repair sequence and scheduling logic are difficult to maintain. Particularly, under the concurrent conditions of communication disconnection, unreachable road and power interruption, the load recovery range of the power distribution network is continuously enlarged, the key load power supply capacity is obviously reduced, and higher requirements are put forward on the toughness of the system. Therefore, the power distribution system not only needs to have the capability of quick restoration after disaster, but also needs to use a scientific and cooperative resource pre-deployment strategy in the pre-disaster stage to improve the survivability and recovery efficiency of the system under disaster impact. In the prior art, the planning and operation of a power distribution network in a disaster scene are mostly based on single network assumption, the system modeling of a cascading failure mechanism among the power distribution network, a traffic network and a communication network is lacked, and meanwhile, the influence of communication link interruption on the processes of rush repair scheduling, network reconstruction, source load control and the like is generally ignored. Studies have focused on planning of repair team paths, temporary power supply of power supplies, drainage operation or physical equipment reinforcement in post-disaster stages, but have focused less on collaborative deployment, accessibility analysis and robustness assessment of pre-disaster multi-type emergency resources (including mobile energy storage, distributed power supplies, emergency repair teams, drainage equipment, unmanned aerial vehicle communication relay platforms, etc.). On the other hand, the traditional intelligent optimization method, such as a particle swarm algorithm, a genetic algorithm and the like, has the defects of low convergence speed, weak global searching capability, easy sinking into local optima and the like when facing the special high-dimension, strong constraint, strong coupling and multi-scene decision space in the pre-disaster deployment problem, and is difficult to cope with complex optimization tasks of huge disaster scene combination, various resource types and constraint chains crossing multiple networks. More importantly, when a communication network is degraded or large-area interrupted due to disasters, the power distribution network often cannot rely on the traditional ground communication infrastructure to realize equipment remote control, scheduling instruction issuing and operation coordination. The key actions such as rush repair path planning, mobile energy storage mobilization, emergency power supply access and the like are limited. Therefore, in the disaster context, the Unmanned Aerial Vehicle (UAV) communication relay platform is used as an emergency communication means which can be rapidly deployed, can be movably covered and has low dependence on ground infrastructure, and becomes a key component for constructing a power dispatching chain in and after the disaster. Communication recovery is achieved through unmanned aerial vehicle relay, controllability of power distribution equipment and authenticity and accessibility of rush-repair team scheduling can be remarkably improved, and therefore overall efficiency of three-network collaborative operation in a disaster scene is improved. In view of the comprehensive, the pre-disaster planning is urgent to introduce an integrated optimization method which takes into account multi-type emergency resource coordination, multi-network coupling, multi-disaster scene and multi-stage operation logic, and simultaneously integrate the enhancement effect of unmanned aerial vehicle communication relay on post-disaster emergency response chains, rely on a more robust intelligent optimization algorithm and a finer three-network physical constraint model, and realize resource optimization configuration and recovery scheduling of pre-disaster, in-disaster and post-disaster three-stage coordination. However, in the existing multi-type resource layout and optimization, the optimization algo