A pre-disaster and in-disaster power supply guarantee collaborative optimization method for distribution network under rain flood disaster

Abstract

This invention discloses a collaborative optimization method for power supply guarantee of distribution networks before and during floods, comprising: obtaining the water depth change process of each distribution node; introducing drainage vehicles to proactively drain water in advance and dynamically correcting the water depth at the nodes; establishing pre-disaster deployment constraints for drainage vehicles and mobile energy storage vehicles and dynamic scheduling constraints during the disaster; constructing a two-stage collaborative optimization model before and during the disaster, including load reduction, drainage vehicle operation, mobile energy storage vehicle operation, distributed power output, and distribution network power flow and network reconfiguration, and transforming the model into a mixed integer second-order cone programming problem for solution. This invention fully utilizes the process characteristic of "gradual water accumulation" in floods, avoids the shutdown of critical equipment by proactively draining water in advance, and combines mobile energy storage, distributed power sources, and network reconfiguration to achieve collaborative optimization of power supply guarantee of the distribution network throughout the entire process of floods, effectively reducing load loss and shortening the duration of power outages.

Description

Technical Field

[0001] This invention relates to the field of power distribution technology, and in particular to a collaborative optimization method for power supply protection of power distribution networks before and during rainstorm disasters. Background Technology

[0002] With the acceleration of global climate change and urbanization, extreme rainstorm events are becoming more frequent and severe, leading to a continuous rise in the risk of urban flooding. As a crucial urban infrastructure, the power distribution network not only supplies electricity to residents and businesses but also provides energy support for vital urban systems such as water supply, communications, and transportation. Failure during rainstorms can easily trigger widespread power outages and secondary impacts. In recent years, numerous incidents of power facility damage caused by rainstorms have occurred both domestically and internationally. For example, in July 2021, Zhengzhou, Henan Province, experienced a severe rainstorm, flooding the underground spaces of more than half of the city's residential areas and important public facilities, damaging 12,425 transformer substations and affecting 1.2663 million users. In September 2023, Shenzhen, Guangdong Province, experienced a severe rainstorm, resulting in severe flooding of numerous substations and power outages in 163 residential communities.

[0003] Outdoor equipment commonly found in power distribution networks, such as prefabricated substations and ring main units, typically includes busbars, switches, and cable connection units. Critical components like cable trays and bottom-entry cables are often located at the bottom of the equipment, making them susceptible to water ingress and moisture damage under flood conditions. When the water depth in the area exceeds the equipment's flood protection height, the equipment will be submerged and shut down, resulting in significant load losses. Existing research mainly focuses on two aspects: one is assessing the flood risk of power distribution equipment, system operational risks, and resilience levels under rainstorm disasters based on two-dimensional hydrodynamic models, flood risk models, or component outage probability models; the other is restoring loads after equipment has already been submerged and shut down, or after a power outage has occurred in the network, by integrating various measures such as emergency repair teams, mobile energy storage, and network reconfiguration.

[0004] Repairing power distribution equipment during rainstorm disasters requires prior drainage, silt removal, and drying, resulting in a lengthy recovery time. Furthermore, rainstorm disasters differ from earthquakes in that they are not instantaneous; the process from flooding to inundation takes time. This time can be utilized for proactive drainage to prevent critical equipment outages and reduce load losses. Therefore, it is necessary to propose a collaborative optimization method for power supply protection in the distribution network throughout the entire rainstorm disaster process. This method, through proactive early drainage and joint scheduling of various emergency resources, aims to minimize critical equipment outages and reduce load losses during rainstorm-induced disasters. Summary of the Invention

[0005] The purpose of this invention is to provide a collaborative optimization method for power supply guarantee of power distribution networks before and during floods. Addressing the problems of power equipment outages and traffic disruptions caused by flooding, this method considers the impact of proactive drainage on changes in water depth. It combines various measures such as drainage vehicles, mobile energy storage vehicles (MESS), distributed generation (DG), and network reconfiguration to pre-deploy resources before a disaster and implement dynamic scheduling during the disaster to reduce load loss, shorten outage duration, and improve power supply guarantee capabilities.

[0006] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0007] This invention includes the following steps:

[0008] Step 1: Simulate the flooding process based on a two-dimensional hydrodynamic model to obtain the change of water depth in the area where each power distribution node is located over time. Introduce the drainage volume and effective drainage coefficient of the drainage truck to dynamically correct the water depth at the node, so as to simulate the impact of proactive early drainage on the evolution of water accumulation and equipment status.

[0009] Step 2: In the pre-disaster phase, based on the docking conditions of each node and the possible degree of disaster, the drainage vehicles and mobile energy storage vehicles are initially pre-deployed; in the disaster phase, the drainage vehicles and mobile energy storage vehicles are dynamically dispatched based on the changes in the equivalent travel time caused by road flooding.

[0010] Step 3: Establish a two-stage collaborative optimization model for power supply guarantee of the distribution network before and during a disaster, with the objective function being the minimum load loss during the disaster, and the decision variables being the load reduction amount, the location and drainage volume of the drainage truck, the location and discharge power of the mobile energy storage vehicle, the output of distributed power sources, and the network topology status. The model comprehensively considers the constraints of load reduction, drainage truck operation, mobile energy storage vehicle operation, and distribution network power flow and network reconfiguration. The nonlinear part of the model is relaxed and transformed into a mixed integer second-order cone programming problem for solution.

[0011] Through the above technical solutions, the present invention can proactively drain water before the power distribution equipment is submerged by floodwater, thereby avoiding the shutdown of critical nodes as much as possible. Under conditions of local island operation or limited support from the main grid, it can further improve the power supply guarantee capability and reduce load loss during disasters by utilizing mobile energy storage, distributed power sources and network reconfiguration.

[0012] The beneficial effects of this invention are:

[0013] By fully utilizing the process characteristics of "gradual development of water accumulation" in rain and flood disasters, we are no longer limited to the passive recovery of power distribution equipment after it is submerged. Instead, we actively drain water in advance through drainage trucks to avoid the shutdown of critical equipment before it is submerged.

[0014] By considering the impact of road flooding on the mobility of mobile resources, we can more realistically establish pre-disaster deployment and dynamic scheduling constraints of drainage vehicles and MESS under rain and flood disasters.

[0015] By integrating changes in water depth, scheduling of drainage vehicles and mobile energy storage, output of distributed power sources and network reconfiguration into the same optimization framework, a comprehensive power supply guarantee method that coordinates pre-disaster and disaster response is formed. Through case studies, this invention verifies that it can reduce the load loss of the distribution network during rain and floods, shorten the total downtime of nodes, and reduce the maximum number of nodes that can be shut down simultaneously. Attached Figure Description

[0016] Figure 1 This is a flowchart of the collaborative optimization method for power supply guarantee of distribution network before and during rainstorm disasters in this invention;

[0017] Figure 2 This is a schematic diagram showing the flow direction of the accumulated water in step 11;

[0018] Figure 3 This is the geographical wiring diagram of the power distribution network system in step 11;

[0019] Figure 4 This is a power distribution network topology diagram of the present invention;

[0020] Figure 5 This is a rainfall scene curve diagram of the present invention;

[0021] Figure 6 This is a graph showing the water accumulation depth at a distribution network node according to the present invention.

[0022] Figure 7 This is a diagram illustrating the mobile resource scheduling process of the present invention under the improved IEEE 33-node power distribution system;

[0023] Figure 8 This is a comparison diagram of the results of the present invention with existing methods under the scenario of a rainstorm with a central rain peak;

[0024] Figure 9 This is a comparison diagram of the results of the present invention with existing methods under the scenario of rain peak leading the rainstorm;

[0025] Figure 10 This is a diagram showing the mobile resource scheduling results under the scenario of a rainstorm with a central rain peak, as described in this invention.

[0026] Figure 11 This is a diagram showing the results of mobile resource scheduling under the scenario of rain peak in front of the rainstorm, according to the present invention. Detailed Implementation

[0027] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0028] like Figures 1-3 As shown: The method provided by this invention mainly includes the following steps:

[0029] Step 1: Based on the results of two-dimensional hydrodynamic analysis, the change process of water depth in the area where each node is located over time is obtained, and the flood prevention height of the equipment is combined to determine whether the node equipment is submerged and shut down; the drainage volume and effective drainage coefficient of the drainage truck are introduced to dynamically correct the water depth of the node, thereby simulating the impact of proactive early drainage on the evolution of water accumulation and equipment status.

[0030] Step 1 specifically includes the following steps:

[0031] Step 11: When flooding causes water accumulation, a schematic diagram of the water flow direction in the grid where the power distribution node is located is shown below. Figure 2 As shown. When considering the water depth in the grid, only the water flow in the four adjacent directions is considered. It is assumed that the water flow is in equilibrium in the vertical direction, while neglecting its horizontal acceleration. The water flow equation obtained from the conservation of momentum is: In the formula, , These are the combined pressure of the surrounding water on the surface water and the frictional resistance of the ground, respectively. For calculating time periods; For water quality; This represents the velocity of the accumulated water. , In the formula, The density of water; It is the acceleration due to gravity; The water depth changes over time; The area is the grid area. The water surface elevation is the sum of the water depth and the terrain height. The conversion factor is the unit. Let be the Manning coefficient. Assuming that the rainfall intensity q at the time of the disaster can be known through meteorological disaster warnings before the disaster occurs, then during the time period... The mass of water falling within the grid is Finally, the information about the water flow velocity was obtained. The quadratic equation in one variable is The water flow velocity can be obtained by removing negative values ​​from the solution. Considering the impact of building cover on surface runoff, the water flow rate of adjacent grids can be calculated as follows: In the formula, This represents the runoff volume along the * direction; a positive value indicates water inflow, and a negative value indicates outflow. Building coverage within the grid; The velocity of the accumulated water along the * direction; The runoff width along the * direction is the grid width in this invention. Simultaneously, considering the reduction in urban drainage systems and surface water due to infiltration and evaporation, this is treated as a negative runoff volume. The natural decrease in water level at time t can be obtained. The water depth in the grid where the power distribution equipment is located is Urban drainage capacity is a key parameter for flood disasters, but accurate information on underground pipe networks is often difficult to obtain. Referring to the "Outdoor Drainage Design Standard," the water accumulation result obtained with a 3-year return period is used as the maximum drainage capacity of the drainage network. The HEC-RAS two-dimensional module is used to discretize and solve the above-mentioned control equations, obtaining the water accumulation depth output of each computational grid at each time period, which serves as input data for the subsequent pre-disaster and in-disaster collaborative optimization model of the power distribution network.

[0032] Step 12: When the water depth in the area where the power distribution equipment is located exceeds the equipment's flood protection height, the power distribution equipment will be submerged and shut down. In the formula, Let be the water depth of the grid containing node i at time t; The flood protection height for the power distribution equipment at node i; Let be a binary variable indicating whether equipment is shut down. A value of 1 indicates that the power distribution equipment at node i is submerged and shut down at time t, and a value of 0 indicates that the equipment is operating normally. Rain and flood disasters often occur over a period of time; the accumulation of water to submersion is not instantaneous. Therefore, drainage trucks can be introduced during this process to proactively drain water in advance. Dispatching drainage trucks to remove water before it exceeds the flood protection height of the power distribution equipment can prevent critical equipment from being submerged and shut down, ensuring power supply during disasters.

[0033] Step 13: After introducing drainage trucks, the change in water depth is no longer solely determined by exogenous factors such as rainfall, but becomes a process that can be actively controlled. Inundation simulations can reveal the changes in water depth in the study area under flood disasters. At this point, by superimposing the drainage volume of the drainage trucks into the water depth formula, a method for describing the change in water depth considering advance drainage can be obtained. The solution principle of HEC-RAS is based on two-dimensional hydrodynamic equations and can be combined with ArcGIS software to simulate urban flooding depth. Without drainage trucks, the change in water depth satisfies: , In the formula, Let be the water depth at node i at time t, as simulated by HEC-RAS. This represents the increase in water accumulation caused by the combined effects of rainfall and surface runoff. When introducing drainage trucks, the effective drainage coefficient is considered. The change in water depth at node i can be represented as In the formula, For the assembly of drainage vehicles; Let be the drainage volume of the k-th drainage vehicle during the time interval from t-1 to t; Let be the water accumulation area at node i. The drainage coefficient is used to equivalently describe the net reduction effect of drainage operations on the water depth of a node, thus converting the drainage volume into a water depth reduction. Within a two-dimensional connected catchment area, although local pumping can directly reduce the water volume of that unit, it will cause backflow due to the surrounding water level difference, resulting in an actual water depth reduction that is less than the theoretical pumping reduction. Therefore, the parameter... To characterize the extent to which this backfilling effect reduces drainage efficiency, we can perform hydrodynamic simulations based on a small number of scenarios involving pumping. Offline calibration.

[0034] Step 2: In the pre-disaster phase, based on the docking conditions of each node and the possible degree of disaster, the drainage vehicles and MESS are initially pre-deployed; in the disaster phase, the drainage vehicles and MESS are dynamically scheduled according to the changes in the equivalent travel time caused by road flooding.

[0035] Step 2 specifically includes the following steps:

[0036] Step 21: Before the onset of heavy rain or before severe flooding occurs on roads, mobile resources can be pre-deployed based on rainfall forecasts and the distribution of flood-prone areas. This invention uses the pre-deployment location as the initial location for dynamic dispatching during disasters. The constraints for the pre-deployment of drainage vehicles are as follows: , , , In the formula, For distribution network nodes; Choose a variable for the starting point of the k-th drainage vehicle, with a value of 1 indicating that the drainage vehicle is located at node i at the initial time; A binary variable is assigned to the drainage vehicle, with a value of 1 indicating that the k-th drainage vehicle is stationed at node i at time t; Let be the number of drainage vehicles that node i can dock at the initial moment. Furthermore, MESS also satisfies the aforementioned pre-disaster deployment constraints.

[0037] Step 22: During the disaster phase, the drainage vehicle and MESS need to move along the road from the pre-deployed nodes to the disaster-stricken nodes to carry out operations. Their arrival time is affected by both their initial location and the degradation of road capacity caused by the disaster. To characterize the impact of road conditions on the movement process, this invention uses an equivalent travel time model to describe the travel time between nodes. The travel time of drainage vehicle k from node i to node j at time t can be expressed as: In the formula, Let k be the travel time of the drainage vehicle from node i to node j; Let be the equivalent travel distance between node i and node j; This represents the equivalent travel speed of the flood drainage truck. Considering the impact of road flooding and traffic congestion on traffic conditions, a disaster impact coefficient is introduced to correct for distance and speed. , In the formula, Let i be the road distance between node i and node j; This represents the depth of water accumulation on road ij at time t; This is a moderating coefficient for the impact of road waterlogging on traffic capacity, used to reflect the degree of degradation of road traffic capacity under rain and flood disaster conditions; The speed of the drainage truck; These are disaster impact factors used to characterize the combined effects of rainfall intensity, road flooding, and traffic flow on the driving capacity of drainage vehicles. Based on the above travel time modeling, the scheduling sequence of drainage vehicles must satisfy the following spatiotemporal constraints: before the drainage vehicle moves from node i to node j, i.e. At that time, its connection state with node j Always 0, that is Furthermore, MESS also satisfies the aforementioned dynamic scheduling constraints during disasters.

[0038] Step 3: Considering the impact of equipment submersion and shutdown during rainstorms and the effects of drainage trucks on changes in water depth, a collaborative optimization model for power supply guarantee of the distribution network before and during disasters is established. The proposed model uses load reduction, the location and operational status of mobile resources, and network topology status as decision variables, with the objective function being the minimum load loss during a disaster. It considers load reduction constraints, pre-disaster deployment constraints and operational constraints of MESS (Mechanical, Energy, and Resources) for dynamic scheduling during disasters, pre-disaster deployment constraints and operational constraints of drainage trucks for dynamic scheduling during disasters, and distribution network operational constraints. After relaxing the nonlinear parts of the model using the Big M method, the model can be transformed into a mixed-integer second-order cone programming problem, which can be solved directly using the Gurobi solver.

[0039] Step 3 specifically includes the following steps:

[0040] Step 31: The core objective of the pre-disaster and during-disaster power supply guarantee collaborative optimization model for distribution networks under rain and flood disasters is to minimize the impact of disasters on distribution network operation by rationally utilizing emergency resources, thereby ensuring power supply during disasters. Based on the above considerations, the objective function of this model is: The cost items are as follows: , , , , In the formula, For the period of disaster; , These are the DG and MESS sets, respectively. For time intervals; Let be the active load that node i is reduced at time t; The active power output of the upstream power grid at time t; The active power output of the g-th DG at time t; Let be the active power of the discharge of the vth MESS at node i at time t; Let be the drainage volume of the k-th drainage vehicle at time t at node i; The unit cost coefficient for load loss. , , , These are the unit cost coefficients for power purchase from the upper-level power grid, DG generation, MESS discharge, and drainage operations, respectively. Under the circumstances of rain and flood disasters, various power supply guarantee measures are coordinated to minimize load loss; therefore, the unit cost coefficient for load loss is set higher than the unit cost coefficients of each of the various measures.

[0041] Step 32: The pre-disaster and during-disaster power supply guarantee collaborative optimization model must satisfy load reduction constraints, MESS and drainage vehicle scheduling constraints, and distribution network operation constraints. All loads at nodes where equipment is out of service are reduced. For nodes where equipment is operating normally, if power supply is insufficient due to a disaster, the load reduction amount must not exceed the initial demand. Assuming a fixed power factor, the load reduction constraint can be expressed as follows: , , In the formula, , These represent the active and reactive loads at node i during normal operation at time t; Let be the reactive load that node i is reduced at time t. The scheduling of the MESS must satisfy pre-disaster deployment and dynamic scheduling constraints during the disaster. Meanwhile, considering the widespread power shortages and equipment submersion shutdowns in disaster-stricken areas, it is difficult for the MESS to obtain reliable power for charging during the disaster phase in actual operation. Furthermore, the timescale of the rainstorm disaster considered in this invention is relatively short, and the initial state of charge (SOC) is sufficient to meet emergency support needs. Therefore, it is assumed that the MESS only performs discharging during the disaster, and the charging process is not considered. Based on this, the MESS also needs to satisfy... , , , , , In the formula, Let t represent whether the vth MESS is located at node i at time t. A value of 1 indicates that it is located at the node, and a value of 0 indicates that it is not located at the node. and Let V be the active power and reactive power of the vth MESS discharged at node i at time t, respectively. and The upper limits of active and reactive power for the discharge of the vth MESS vehicle; Let the SOC size of the vth MESS at time t; and Let the upper and lower limits of the SOC of the vth MESS be defined. This is the maximum capacity for MESS. For MESS discharge efficiency; This refers to the safe water level that MESS can drive into. For a large M value. In addition to satisfying pre-disaster deployment and dynamic scheduling constraints during disasters, the dispatching of drainage vehicles also needs to satisfy constraints on location uniqueness, pumping capacity limitations, and drainage rationality, specifically expressed as follows: , , , , , , In the formula, This is an indicator variable for whether drainage should begin. A value of 1 indicates that the drainage truck should start draining water, while a value of 0 indicates that it should not. For drainage vehicles in a single time period Maximum internal drainage capacity; This represents the upper limit of the drainage capacity of flood control vehicles during disasters; This is an indicator variable for whether a node has water accumulation. A value of 1 indicates that node i has water accumulation at time t, and vice versa. This is a very small number. For radial distribution networks, the DistFlow power flow equations are used, and the Big-M method is employed to relax the voltage equations for disconnectable lines. Furthermore, this invention also considers distribution network topology reconfiguration measures, so reconfiguration constraints must be satisfied to avoid the formation of loops.

[0042] After the above three steps, a collaborative optimization model for power supply guarantee of the power distribution network before and during rainstorm disasters can be obtained.

[0043] Figure 3 This is an elevation map obtained by mapping the IEEE-33 node power distribution system to a local area of ​​a city in Southwest China. The darker the color, the lower the terrain of the power distribution node, and the more likely it is to accumulate water during rainstorms. Figure 4 The improved IEEE 33-node distribution network adds five distribution groups (DGs) to nodes 3, 6, 17, 25, and 29 on the basis of the original topology. Figure 5 The rainfall time series curve was obtained by selecting a rainfall peak coefficient of 0.4 using the Chicago rainfall pattern method. Figure 6 Is Figure 5 The rainfall scenario was obtained from HEC-RAS simulation. Figure 3 Water depth curves at various nodes. Figure 7 Is Figure 5The mobile resource dispatch process under the disaster of urban flooding caused by extreme rainstorms: the three drainage vehicles were pre-deployed at nodes 4, 17, and 31, and were... Figure 3 It is known that nodes 2, 4, and 5 form a concentrated flood-prone area. Among them, node 4 has the largest load and is located in the center of the area. Pre-deploying the No. 1 drainage truck here is beneficial for rapid deployment to nodes 2 and 5 during the disaster. If the power supply to node 4 is not prioritized, a large amount of load will be lost, and it will also cause a large-scale power outage on lines 2-4 or 4-5, making the restoration pressure even greater. Node 17 has the lowest terrain and the fastest water accumulation. The No. 2 drainage truck is pre-deployed here to handle some of the water accumulation. This can reduce the difficulty of subsequent repairs at a lower cost in the early stages of the disaster. Then, at t=2, it will switch to the more important node 12. This node has a tie switch connected to it and is located at the front of the main line. Once it is flooded and shut down, the tie line 12-22 will not be able to close, reducing the system's reconfiguration means. At the same time, the available power supply to the downstream section of the main line will be insufficient, resulting in a greater load reduction. Although node 31 is far from the center of the flooding, it carries a significant load. Its shutdown would cause nodes 32 and 33 to also shut down, and the tie line 18-33 would also fail to close. Therefore, drainage truck #3 was deployed here. At t=3, node 14 was flooded, and a single drainage truck could not quickly reduce the water level to a safe range. Therefore, drainage trucks #2 and #3 worked together to clear the floodwater at node 14. Simultaneously, drainage truck #1 addressed the floodwater at node 2 to expedite the restoration of the main power supply channel. Once the main power supply channel was restored, only nodes 5 and 17 required final drainage and maintenance before the system could return to normal. MESS was pre-deployed at nodes 20 and 24 before the disaster. Node 20 is located on the upper branch, and the DG in this area is insufficient; if it becomes an island, it will rely even more heavily on external power. Nodes 24 and 25 are load aggregation areas, and pre-deploying them here maximizes load supply. At t=4, the main grid channel gradually recovers. At this time, the power flow is redistributed. Node 16 lacks DG and is located in a critical area in the latter part of the trunk line. MESS2 goes to node 16 to provide more direct and effective support to this area. When the main grid channel and network structure are basically restored, MESS withdraws its support to avoid ineffective discharge. Figure 8 The results show that both the load loss and the number of outage nodes are significantly less than those of existing methods. The invention resulted in a total load loss of 520.3 kWh, far less than the 1738 kWh of existing methods; the total outage time of the nodes in the invention was 10 hours, far less than the 19 hours of existing methods. Furthermore, the Chicago rainfall pattern method was used, with a peak rainfall coefficient of 0.2, to simulate a leading-peak rainstorm scenario. Figure 9 This is a comparison of the load loss situation of the present invention and existing methods under this rainstorm scenario. Similarly, both the load loss amount and the number of outage nodes are significantly less in the present invention than in the existing methods. The present invention has a total load loss of 343 kWh, far less than the 1260 kWh of the existing methods; the nodes of the present invention are out of service for a total of 10 hours, far less than the 19 hours of the existing methods.

[0044] Figures 10~11 This indicates that in the case of a rainstorm with a leading rain peak, the critical relocation time of the drainage trucks and MESS units is advanced by one time step. Drainage truck No. 2, which could have drained some water from node 17 earlier, was instead pre-deployed at node 2 due to time constraints, prioritizing the rapid restoration of the main network channel. Because nodes 2 and 14 are far apart, drainage truck No. 2 abandoned its task of draining water from node 14 along with truck No. 3, and instead went to handle the relatively closer node 5. Since the main network channel is restored earlier in the case of a leading rain peak, both MESS units were pre-deployed at a later position on the main line, and then relocated to the vicinity of the out-of-operation nodes to provide power output when islanding occurs. These results demonstrate that the scheduling strategies of drainage trucks and MESS units are reasonably adjusted under different rainstorm disaster scenarios to adapt to different flooding conditions, further illustrating the applicability of this invention.

[0045] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A collaborative optimization method for power supply guarantee of distribution networks before and during rainstorm disasters, characterized in that, Includes the following steps: Step 1: Simulate the flooding process based on a two-dimensional hydrodynamic model to obtain the change of water depth in the area where each power distribution node is located over time. Introduce the drainage volume and effective drainage coefficient of the drainage truck to dynamically correct the water depth at the node, so as to simulate the impact of proactive early drainage on the evolution of water accumulation and equipment status. Step 2: In the pre-disaster phase, based on the docking conditions of each node and the possible degree of disaster, the drainage vehicles and mobile energy storage vehicles are initially pre-deployed; in the disaster phase, the drainage vehicles and mobile energy storage vehicles are dynamically dispatched based on the changes in the equivalent travel time caused by road flooding. Step 3: Establish a two-stage collaborative optimization model for power supply guarantee of the distribution network before and during a disaster, with the objective function being the minimum load loss during the disaster, and the decision variables being the load reduction amount, the location and drainage volume of the drainage truck, the location and discharge power of the mobile energy storage vehicle, the output of distributed power sources, and the network topology status. The model comprehensively considers the constraints of load reduction, drainage truck operation, mobile energy storage vehicle operation, and distribution network power flow and network reconfiguration. The nonlinear part of the model is relaxed and transformed into a mixed integer second-order cone programming problem for solution.

2. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 1, characterized in that: In step 1, by adding the drainage volume of the flood discharge truck to the formula for calculating water depth, a dynamic change expression for water depth considering advance drainage is obtained: In the formula: Let be the water depth of the grid containing node i at time t. For the assembly of drainage vehicles; Let be the drainage volume of the k-th drainage vehicle during the time interval from t-1 to t; Let be the area of ​​water accumulation at node i. This indicates the increase in water accumulation caused by the combined effects of rainfall and surface runoff.

3. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 1, characterized in that: In step 2, the initial pre-deployment of the drainage vehicle and the mobile energy storage vehicle shall meet the following constraints: ， ， ， In the formula, For distribution network nodes; Choose a variable for the starting point of the k-th drainage vehicle, with a value of 1 indicating that the drainage vehicle is located at node i at the initial time; A binary variable is assigned to the drainage vehicle, with a value of 1 indicating that the k-th drainage vehicle is stationed at node i at time t; Let represent the number of drainage vehicles that can dock at node i at the initial moment.

4. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 3, characterized in that: In step 2, the dynamic scheduling of drainage vehicles and mobile energy storage vehicles during the disaster phase adopts an equivalent travel time model, from node i to node... At any moment The passage time is expressed as: In the formula, Let k be the travel time of the drainage vehicle from node i to node j; Let be the equivalent travel distance between node i and node j; This is the equivalent travel speed of the drainage vehicle.

5. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 1, characterized in that: In step 3, the objective function is expressed as: The cost items are as follows: ， ， ， ， In the formula, For the period of disaster; , These are the DG and MESS sets, respectively. For time intervals; Let be the active load that node i is reduced at time t; The active power output of the upstream power grid at time t; The active power output of the g-th DG at time t; Let be the active power of the discharge of the vth MESS at node i at time t; Let be the drainage volume of the k-th drainage vehicle at time t at node i; The unit cost coefficient for load loss. , , , These are the unit cost coefficients for purchasing electricity from the upper-level power grid, generating DG, discharging MESS, and drainage operations, respectively.

6. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 1, characterized in that: In step 3, the mobile energy storage vehicle only discharges during the disaster and does not charge. Its operational constraints include: ， ， ， ， ， In the formula, Let t represent whether the vth MESS is located at node i at time t. A value of 1 indicates that it is located at the node, and a value of 0 indicates that it is not located at the node. and Let V be the active power and reactive power of the vth MESS discharged at node i at time t, respectively. and The upper limits of active and reactive power for the discharge of the vth MESS vehicle; Let the SOC size of the vth MESS at time t; and Let the upper and lower limits of the SOC of the vth MESS be defined. This is the maximum capacity for MESS. For MESS discharge efficiency; This refers to the safe water level that MESS can drive into. For large M values.

7. The method for coordinated optimization of power supply guarantee for distribution networks before and during rainstorm disasters as described in claim 1, characterized in that: In step 3, the dispatching of drainage vehicles must also meet constraints on location uniqueness, drainage capacity limitation, and drainage rationality, specifically as follows: ， ， ， ， ， ， In the formula, This is an indicator variable for whether drainage should begin. A value of 1 indicates that the drainage truck should start draining water, while a value of 0 indicates that it should not. For drainage vehicles in a single time period Maximum internal drainage capacity; This represents the upper limit of the drainage capacity of flood control vehicles during disasters; This is an indicator variable for whether a node has water accumulation. A value of 1 indicates that node i has water accumulation at time t, and vice versa. It is an extremely small number.

Images