Smart city waterlogging risk prediction methods and IoT systems, devices and media

Abstract

This specification provides a smart city waterlogging risk prediction method, an Internet of Things (IoT) system, device, and medium. The method is implemented based on a management platform of the smart city waterlogging risk prediction IoT system. The method includes: predicting the waterlogging risk of the target area based on the acquired regional information of the target area; determining the corresponding adjustment scheme for the target area based on the waterlogging risk; and executing the adjustment instructions corresponding to the adjustment scheme.

Description

Technical Field

[0001] This specification relates to the fields of Internet of Things and cloud platforms, and in particular to a method for predicting waterlogging risks in smart cities, as well as an Internet of Things system, device, and storage medium. Background Technology

[0002] In modern life, due to the influence of weather and geological disasters, water accumulation often occurs in the area, causing great inconvenience to people's normal lives.

[0003] Therefore, there is a need for a smart city flood risk prediction method and an Internet of Things (IoT) system, device, and storage medium that can determine and implement countermeasures by predicting flood risks, so as to quickly prepare for flooding and avoid casualties and economic losses. Summary of the Invention

[0004] This specification provides one or more embodiments of a smart city flood risk prediction method, implemented based on the management platform of a smart city flood risk prediction IoT system. The method includes: predicting the flood risk of a target area based on acquired regional information; determining a corresponding adjustment scheme for the target area based on the flood risk; and executing adjustment instructions corresponding to the adjustment scheme.

[0005] This specification provides one or more embodiments of a smart city flood risk prediction IoT system. The system includes a management platform, a user platform, a service platform, a sensor network platform, and an object platform. The management platform includes a main database and several management sub-platforms. The sensor network platform includes several sensor network sub-platforms. The area information is obtained based on the object platform and transmitted by the corresponding sensor network sub-platform to the corresponding management sub-platform, and then uploaded by the management sub-platform to the main database. The management platform is configured to perform the following operations: predict the flood risk of the target area based on the obtained area information; determine the corresponding adjustment scheme for the target area based on the flood risk; and execute the adjustment instructions corresponding to the adjustment scheme.

[0006] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the smart city waterlogging risk prediction method as described in any of the above embodiments.

[0007] This specification provides one or more embodiments of a smart city waterlogging risk prediction device, the device including at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least some of the computer instructions to implement the above-described smart city waterlogging risk prediction method. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is a schematic diagram of a smart city waterlogging risk prediction system according to some embodiments of this specification;

[0010] Figure 2 This is an exemplary flowchart of a smart city waterlogging risk prediction method according to some embodiments of this specification;

[0011] Figure 3-1 This is a schematic diagram illustrating the determination of road flooding risk for each road in a target area during a target time period based on a flooding risk prediction model, according to some embodiments of this specification.

[0012] Figure 3-2 This is a schematic diagram of the regional road network structure data according to some embodiments of this specification;

[0013] Figure 4-1 This is a schematic diagram illustrating the first joint control scheme for determining the target area based on a joint scheduling model, according to some embodiments of this specification.

[0014] Figure 4-2 This is a schematic diagram of the regional road network infrastructure structure data according to some embodiments of this specification;

[0015] Figure 5 This is an exemplary flowchart illustrating a material scheduling scheme for determining a target area, based on some embodiments of this specification. Detailed Implementation

[0016] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0017] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0018] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0020] Figure 1 This is a schematic diagram of a smart city flood risk prediction system 100 according to some embodiments of this specification.

[0021] It should be understood that a smart city flooding risk prediction system 100 can be implemented in various ways. For example... Figure 1 As shown, the smart city flooding risk prediction system includes a user platform, a service platform, a management platform, a sensor network platform, and an object platform. In some embodiments, the smart city flooding risk prediction system may be part of or implemented by the processing equipment.

[0022] A user platform can refer to a user-centric platform. For example, a user platform can obtain user input commands through a terminal to query flood risks and mitigation plans. Alternatively, a user platform can provide feedback on mitigation plans to users.

[0023] A service platform refers to a platform that provides input and output services to users. For example, a service platform can transmit adjustment plans to user platforms.

[0024] The management platform can refer to the platform that coordinates and integrates the connections and collaborations between various functional platforms, and gathers all the information from the smart city waterlogging risk prediction system. The management platform can provide perception management and control management functions for the operation of the smart city waterlogging risk prediction system.

[0025] In some embodiments, the management platform may include a central database and several sub-platforms. One region may correspond to one sub-platform. For example, a sub-platform may upload regional information within the target region obtained through the object platform to the central database of the management platform. The central database of the management platform may, based on the regional information, predict the flooding risk of the target region; determine the corresponding adjustment plan based on the flooding risk; execute the adjustment instructions corresponding to the adjustment plan; and transmit the adjustment plan to the user platform through the service platform. In some embodiments, the management platform may be a remote platform controlled by administrators, artificial intelligence, or preset rules.

[0026] The management platform can be configured to: obtain regional information of the target area through the object platform; determine the water accumulation risk of the target area based on the regional information; determine the corresponding adjustment plan for the target area based on the water accumulation risk; and execute the adjustment instructions corresponding to the adjustment plan. For more information on executing the adjustment instructions corresponding to the adjustment plan, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0027] In some embodiments, the management platform is configured to further process regional information based on a joint scheduling model to determine a first joint control scheme corresponding to the target region; the joint scheduling model is a machine learning model. For more information on determining the first joint control scheme and the joint scheduling model, please refer to [link to relevant documentation]. Figure 4-1 , Figure 4-2 And its related descriptions.

[0028] In some embodiments, the management platform is configured to further acquire an initial material dispatch plan, which includes the initial supply of relief materials from each rescue point to each demand point; iteratively update the initial material dispatch plan based on a first preset algorithm to obtain at least one candidate material dispatch plan; calculate an evaluation value for each candidate material dispatch plan based on a second preset algorithm; and use the candidate material dispatch plan whose evaluation value meets preset requirements as the initial material dispatch plan for the next iteration, iterating until the iteration completion condition is met. For more information on determining the material dispatch plan corresponding to the target area through iteration, please refer to [link to relevant documentation]. Figure 5 And its related descriptions.

[0029] A sensor network platform can refer to a functional platform that manages sensor communication. In some embodiments, the sensor network platform can connect a management platform and an object platform to realize the functions of sensing communication for both perception and control information. In some embodiments, the sensor network platform may include several sensor network sub-platforms. A region may correspond to one sensor network sub-platform and one management sub-platform. For example, a sensor network sub-platform can transmit region information within the corresponding target region obtained through the object platform to the corresponding management sub-platform.

[0030] An object platform can refer to a functional platform for generating perceived information. In some embodiments, the object platform can acquire information, such as regional information about a target area.

[0031] In some embodiments, the smart city flood risk prediction system can be applied to various scenarios for flood risk prediction. In some embodiments, the system can acquire relevant information for flood risk prediction in a target area under various scenarios (e.g., road network information, precipitation information, drainage capacity information, historical flood information, etc.) to obtain the flood risk under each scenario. In some embodiments, based on the acquired flood risks under each scenario, the system can derive adjustment schemes for addressing flood risks in each scenario (e.g., joint scheduling schemes for water storage facilities in the target area, material scheduling schemes, etc.).

[0032] This specification also provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes any of the smart city flooding risk prediction methods provided in this specification.

[0033] It should be noted that the above description of the smart city flooding risk prediction system and its modules is for convenience only and should not be construed as limiting this specification to the scope of the illustrated embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. For example, Figure 1 The management platform, user platform, service platform, sensor network platform, and object platform disclosed in the document can be different platforms within the same system, or a single platform can implement the functions of two or more of the aforementioned platforms.

[0034] Figure 2 This is an exemplary flowchart illustrating a smart city waterlogging risk prediction method according to some embodiments of this specification. Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by a management platform.

[0035] Step 210: The management platform can predict the risk of water accumulation in the target area based on the acquired regional information of the target area.

[0036] The target area can refer to areas where flooding has already occurred or may occur. For example, the target area could be an area currently experiencing heavy rainfall, surrounding areas affected by heavy rainfall, areas flooded due to upstream flooding, typhoon landfall areas, or areas experiencing seawater intrusion.

[0037] Regional information can refer to information related to waterlogging prediction that reflects the target area.

[0038] In some embodiments, the regional information may include road network information, precipitation information, drainage capacity information, and historical waterlogging information for the target area. Further details regarding the above information can be found in [link to relevant documentation]. Figure 3-1 , Figure 3-2 And its related descriptions.

[0039] In some embodiments, the area information may further include information on the deployment of water storage facilities in the target area, the water storage capacity of the facilities, the drainage capacity of the facilities, and the opening and closing information of the facilities. For more details on the above information, please refer to [link to relevant documentation]. Figure 4-1 , Figure 4-2 And its related descriptions.

[0040] In some embodiments, the regional information may further include target area delineation information, demand information at demand points, waterlogging risk at demand points, priority coefficients for demand points, delineation information of rescue points within the target area, material information at rescue points, and historical rescue information for rescue points. For further explanation of the above information, please refer to [link to relevant documentation]. Figure 5 And its related descriptions.

[0041] Flooding risk refers to the likelihood of flooding in a target area. In some embodiments, flooding risk can be characterized as a flooding risk level. A higher flooding risk level indicates a greater likelihood of flooding in the target area and a larger volume of flooding. As an example only, the flooding risk level can be determined based on the water level in the target area. For instance, the management platform can preset a water level threshold. The water level threshold may include a first water level threshold and a second water level threshold, wherein the second water level threshold is higher than the first water level threshold. When the water level in the target area is less than the first water level threshold, the flooding risk level of the target area can be level 1. When the water level in the target area is greater than or equal to the first water level threshold but less than the second water level threshold, the flooding risk level of the target area can be level 2. When the water level in the target area is greater than the second water level threshold, the flooding risk level of the target area can be level 3.

[0042] In some embodiments, the management platform can perform modeling or employ various data analysis algorithms to analyze and process the regional information of the target area to determine the flooding risk of the target area. As an example only, the management platform can pre-determine historical regional information vectors corresponding to historical regional information, and generate a flooding risk mapping table based on the historical actual flooding risk corresponding to each historical regional information vector. Understandably, the management platform can determine the predicted regional information vector corresponding to the current regional information based on the current regional information. Further, the management platform can determine at least one target regional information vector from the historical regional information vectors based on the vector distance (e.g., Euclidean distance) between the predicted regional information vector and the historical regional information vectors. The management platform can use the flooding risk corresponding to at least one target regional information vector as the current flooding risk.

[0043] In some embodiments, the management platform can determine the road flooding risk of each road in a target area during a target time period based on the processing of regional information using a flooding risk prediction model, wherein the flooding risk prediction model is a machine learning model. Further details regarding the above embodiments can be found in [link to relevant documentation]. Figure 3-1 , Figure 3-2 And its related descriptions.

[0044] Step 220: The management platform can determine the corresponding adjustment plan for the target area based on the risk of water accumulation.

[0045] A mitigation plan can refer to the relief and rescue measures taken to address the risk of flooding. For example, a mitigation plan could be the joint operation of drainage systems. Another example is the scheduling and optimization of flood control equipment and materials. Yet another example is flood control using underground spaces.

[0046] In some embodiments, the management platform can pre-set adjustment plans corresponding to various waterlogging risks. The platform can then determine the corresponding adjustment plan for the target area based on the predicted relationship between waterlogging risks and adjustment plans. For example, when the waterlogging risk is level 1, the corresponding adjustment plan could be to activate the drainage system and water storage facilities. When the waterlogging risk is level 2, the corresponding adjustment plan could be to activate the drainage system and water storage facilities while preparing specialized flood control equipment and emergency supplies for urban flooding. When the waterlogging risk is level 3, the corresponding adjustment plan could be to activate the drainage system and water storage facilities, prepare specialized flood control equipment and emergency supplies for urban flooding, while simultaneously evacuating people from the target area and conducting underground space rescue operations.

[0047] In some embodiments, in response to a flood risk meeting a first preset condition, the management platform can determine a first joint control scheme corresponding to the target area based on regional information. The first joint control scheme includes a joint scheduling scheme for the water storage facilities in the target area. Further details regarding the above embodiments can be found in [link to relevant documentation]. Figure 4-1 , Figure 4-2 And its related descriptions.

[0048] In some embodiments, in response to the water accumulation risk meeting a second preset condition, the management platform can determine a material dispatching plan corresponding to the target area based on regional information. Further details regarding the above embodiments can be found in [link to relevant documentation]. Figure 5 And its related descriptions.

[0049] Step 230: The management platform can execute the adjustment instructions corresponding to the adjustment plan.

[0050] A regulation instruction refers to an instruction used to implement a regulation scheme. In some embodiments, the management platform can pre-set one or more regulation instructions corresponding to a regulation scheme to form a mapping relationship. For example, when the regulation scheme is the joint scheduling of a drainage system, the regulation instruction may be to activate the drainage system and storage facilities. When the regulation scheme is the scheduling and optimization of flood control equipment and materials, the regulation instruction may be to dispatch personnel to prepare flood control equipment and emergency supplies. When the regulation scheme is flood control in underground spaces, the regulation instruction may be emergency evacuation of passengers, infrastructure inspection, organization of emergency operations, etc.

[0051] In some embodiments, the management platform can execute regulation commands by issuing control commands. For example, the management platform can directly send control commands to the storage facilities within the target area to activate or deactivate them. Another example is that the management platform can directly send control commands to the material dispatch system within the target area to dispatch flood control equipment and relief supplies.

[0052] In some embodiments, the management platform can execute control commands by issuing prompts. For example, the management platform can directly send prompts to personnel within a target area via SMS, text push, images, videos, voice, broadcasts, etc., to achieve emergency evacuation of personnel within the area. Alternatively, the management platform can first identify the target terminal of the personnel or department directly executing the control command and send the prompts to that target terminal via telephone or SMS. After receiving the prompts, relevant personnel can activate drainage systems and storage facilities, prepare flood control equipment and rescue supplies, urgently evacuate passengers, inspect infrastructure, and organize emergency operations.

[0053] The methods described in some embodiments of this specification predict the risk of water accumulation in the target area based on the area information of the target area, thereby determining and implementing the corresponding adjustment plan. This can respond to the risk of water accumulation in advance and quickly, avoiding casualties and economic losses.

[0054] In some embodiments, the regional information may include: road network information, precipitation information, drainage capacity information, and historical water accumulation information of the target region.

[0055] Road network information can be data describing the road network of a target area. For example, road network information can include information about each intersection in the target area (such as intersection location, altitude, intersection type, etc.) and information about each road (such as intersections at both ends of the road, road width, road length, and road slope, etc.).

[0056] Precipitation information reflects the precipitation situation in a target area within a target time period. For example, precipitation information can include the amount of precipitation in the target area caused by weather such as rain, snow, and hail within the target time period.

[0057] Drainage capacity information can refer to the ability to discharge liquid water on the ground surface of a target area. For example, drainage capacity information may include the maximum amount of water that the target area can discharge per unit time, such as 10,000 cubic meters per hour or 300,000 cubic meters per day. In some embodiments, drainage capacity information may be related to the opening and closing operation information of storage facilities to which the water flows from the target area. For example, when a storage facility used for flood discharge in the target area is opened, the drainage capacity of the target area is enhanced. Further details regarding facility opening and closing operation information can be found in this specification. Figure 4-1 , Figure 4-2 And its related descriptions.

[0058] Historical flooding information refers to the flooding situation of a target area over a historical period. In some embodiments, historical flooding information can be represented by water level, such as 0.7 meters, 1.5 meters, etc. In some embodiments, historical flooding information can also reflect the degree of congestion at various intersections in the target area. For example, if an intersection has experienced flooding multiple times in the past, and very frequently, then that intersection is more prone to congestion.

[0059] In some embodiments, the management platform can determine the road flooding risk of each road in the target area during a target time period based on the processing of regional information using a flooding risk prediction model. The flooding risk prediction model is a machine learning model.

[0060] The target time period can be the time period during which the risk of flooding needs to be predicted. The target time period can be any length of time that includes the past, present, or future. For example, the next 12 hours, the next week, etc.

[0061] In some embodiments, the target time period may include at least a portion of the duration of precipitation in the target area. For example, the target time period may be the time from the start of the rainstorm to its cessation in the target area. In some embodiments, the target time period may include multiple preset sub-time periods. For example, the target time period may include the first hour, the second hour, ... the nth hour after the start of the rainstorm.

[0062] In some embodiments, the target time period can be determined based on precipitation conditions and actual needs. For example, if a user anticipates that a current rainstorm may cause flooding in a target area, the target time period can begin when the rainfall rate exceeds the drainage capacity of the target area and end at a preset time or the estimated end time of the rainstorm. As another example, when a user is statistically analyzing the flooding situation in a target area due to snowfall, the target time period can begin when the snow has not melted naturally and end when the snow melts naturally.

[0063] Road flooding risk can be used to describe the probability of flooding on various roads within a target area during a target time period. In some embodiments, road flooding risk can be characterized as a road flooding risk level. Similar to the aforementioned flooding risk, a higher road flooding risk level indicates a greater risk of flooding on the road and a larger amount of flooding is expected.

[0064] Figure 3-1 This is a schematic diagram illustrating the determination of road flooding risk for each road in a target area during a target time period based on a flooding risk prediction model, according to some embodiments of this specification. For example... Figure 3-1 As shown, the water accumulation risk prediction model 330 can be used to determine the road water accumulation risk 340 of each road in the target area during the target time period. The water accumulation risk prediction model 330 can be a machine learning model.

[0065] In some embodiments, the input to the flood risk prediction model 330 can be the regional information 310 of the target area, and the output of the flood risk prediction model 330 can be the road flood risk 340 of each road in the target area during a target time period. In some embodiments, the flood risk prediction model 330 can be a long short-term memory artificial neural network, a graph neural network model, etc.

[0066] The flood risk prediction model 330 can be obtained through training. Training data can be regional information of the sample area in the first sample time period, and labels can be the actual flood risk of each road in the sample area in the second sample time period. The first sample time period is earlier than the second sample time period. Multiple training samples are input into the initial flood risk prediction model. A loss function is constructed based on the output and labels of the initial flood risk prediction model, and the parameters of the initial flood risk prediction model are iteratively updated based on the loss function. Training ends when the trained model meets preset conditions, and the trained initial flood risk prediction model 330 is obtained. Preset conditions may include, but are not limited to, loss function convergence, loss function value being less than a preset value, or the number of training iterations reaching a threshold.

[0067] In some embodiments, the management platform can determine the regional road network map structure data 320 of the target area based on the regional information 310 of the target area, process the regional road network map structure data 320 based on the water accumulation risk prediction model 330, and determine the road water accumulation risk 340 of each road in the target area during the target time period.

[0068] Figure 3-2 This is a schematic diagram of regional road network map structure data according to some embodiments of this specification. The regional road network map structure data 320 is a data structure composed of nodes and edges, with edges connecting nodes, and nodes and edges can have attributes.

[0069] In some embodiments, nodes may correspond to individual intersections within a target area. For example... Figure 3-2 As shown, the first nodes 1, 2, 3, 4, and 5 in the regional road network map structure data 320 can be matched one-to-one with five actual intersections. Node attributes can reflect the relevant characteristics of the corresponding intersections. In some implementations, node attributes may include the corresponding intersection's primary precipitation characteristics, primary drainage characteristics, and congestion-prone characteristics.

[0070] The first drainage characteristic describes the drainage capacity of each intersection within the target area. This first drainage characteristic can be quantitatively represented by a numerical value, such as the volume of drainage from the intersection per unit time (e.g., 1000 cubic meters per hour).

[0071] The first precipitation feature can describe the precipitation amount at each intersection within the target area during the target time period. In some embodiments, the first precipitation feature can be determined based on precipitation information. In some embodiments, the first precipitation feature may also include the current existing water level at each intersection within the target area, such as 50 cm.

[0072] Congestion level describes the degree of congestion at various intersections within a target area. In some embodiments, congestion level can be determined based on historical flooding information at the intersection. For example, if an intersection has experienced frequent flooding in the past, then the congestion level of that intersection is considered high.

[0073] In some embodiments, the edges may correspond to the roads within the target area.

[0074] In the regional road network map structure data 320, the edges are directed edges, and their direction can be determined based on altitude differences. For example, the edge direction can be from high to low altitude. Figure 3-2 As shown, the first edges a, b, c, d, e, and f in the regional road network map structure data 320 can represent roads between five actual intersections. Among the nodes corresponding to two intersections connected by roads, the edges point from the node corresponding to the intersection with higher elevation to the node corresponding to the intersection with lower elevation. For example, according to the first edge a, the elevation of the first node 1 is higher than that of the first node 2.

[0075] Edge attributes can reflect the relevant characteristics of the corresponding road. In some embodiments, edge attributes may include the second precipitation characteristic, second drainage characteristic, slope, length, etc. of the corresponding road.

[0076] The second drainage characteristic describes the drainage capacity of each road within the target area. This second drainage characteristic can be quantitatively represented by a numerical value, such as the volume of water drained by the road per unit time (e.g., 2000 cubic meters per hour).

[0077] The second precipitation feature can describe the precipitation amount of each road within the target area during the target time period. In some embodiments, the second precipitation feature can be determined based on precipitation information. In some embodiments, the second precipitation feature may also include the current existing water level of each road within the target area, such as 50 cm.

[0078] Gradient describes the steepness of a road's slope corresponding to a directed edge. Gradient can be expressed as a slope angle, such as 20°. Length, on the other hand, describes the distance a road travels between two intersections, such as 3 km.

[0079] like Figure 3-1 As shown, the waterlogging risk prediction model 330 can process the regional road network map structure data 320 to determine the road waterlogging risk 340 of each road in the target area during the target time period.

[0080] In some embodiments, the flood risk prediction model 330 can be a graph neural network (GNN) model. The input to the flood risk prediction model 330 can be the regional road network structure data 320 of the target area, and the output can be the road flood risk 340 of each road in the target area during a target time period. The edge outputs of the graph neural network model correspond to the road flood risk of the road during the target time period. For example, the road flood risk of each road can be represented as a flood risk vector output by each edge. The flood risk vector can include one or more elements. When multiple elements are included, each element corresponds to a different road flood risk level, and the element value represents the probability of being at that road flood risk level. For example, if the flood risk vector output by a certain edge is (0.8), it means that the probability of a flood disaster occurring on the road corresponding to that edge is 80%. The management platform can pre-set that if the road flood depth exceeds a preset depth value (e.g., 60 cm), it is considered a flood disaster. For example, the water accumulation risk vector output by a certain side can be (0.6, 0.2), indicating that the probability of a Level 1 water accumulation risk on that road is 60%, and the probability of a Level 2 water accumulation risk is 20%. The higher the level, the greater the water depth on the road. In some embodiments, the management platform can preset a risk threshold. If the probability value of a water accumulation disaster or the probability of a corresponding level of water accumulation risk output by a certain road exceeds the preset risk threshold, then the road is considered a water accumulation risk section.

[0081] In some embodiments, the management platform can train an initial flood risk prediction model based on training data to determine a flood risk prediction model 330. The training data may include training samples and training labels. The training samples may be historical regional road network map structure data determined based on historical data. The training labels may be determined based on historical road flooding conditions. For example, if the historical flood depth of a road exceeds a preset depth value, the label value of that node is 1; if the historical flood depth is below the preset depth value, the label value of the node is a value between 0 and 1, and the closer the flood depth is to the preset depth value, the closer the label value of the node is to 1. The labeled training samples are input into the initial flood risk prediction model, and the parameters of the initial flood risk prediction model are updated through training. When the trained model meets preset conditions, the training ends, and the trained flood risk prediction model 330 is obtained.

[0082] In some embodiments, the nodes and edges of the historical regional road network map structure data can be the same as or similar to the regional road network map structure data to be predicted, and the specific attributes of the nodes and edges can be determined based on road network information, historical precipitation information, historical drainage capacity information, and historical water accumulation information.

[0083] The methods described in some embodiments of this specification process regional road network map structure data through a water accumulation risk prediction model, and combine the actual flow of water at various intersections and roads to further improve the accuracy of road water accumulation risk prediction.

[0084] The methods described in some embodiments of this specification can effectively predict road waterlogging risk through a waterlogging risk prediction model, thereby enabling faster and more accurate determination of regional waterlogging risk, so as to quickly determine adjustment plans and address waterlogging risk in advance.

[0085] In some embodiments, the area information includes first information and second information about the target area.

[0086] In some embodiments, the first information includes road network information, precipitation information, drainage capacity information, and historical waterlogging information for the target area. For detailed explanations regarding road network information, precipitation information, drainage capacity information, and historical waterlogging information, please refer to [link to relevant documentation]. Figure 3-1 , Figure 3-2 The details and their corresponding descriptions will not be repeated here.

[0087] In some embodiments, the second information includes information on the deployment of water storage facilities in the target area, information on the water storage capacity of the facilities, information on the drainage capacity of the facilities, and information on the opening and closing of the facilities.

[0088] Storage and regulation facilities refer to facilities that receive, store, and discharge rainwater, road floodwater, and other water-related pollutants, such as rainwater storage ponds, intermediate pumping stations, and reservoirs in rivers and lakes. Information on the deployment of storage and regulation facilities can include the type, location, and number of facilities.

[0089] Information on the water storage capacity of a facility may include the maximum water storage capacity of the storage facility or the currently available water storage capacity.

[0090] Facility drainage capacity information can refer to the maximum drainage rate of the storage facility. In some embodiments, facility drainage capacity information can be characterized by the volume of water that can be discharged per unit time.

[0091] Facility opening and closing information refers to the opening and closing status of the inlet and outlet gates of a water storage facility. This information reflects the operational status of the facility. The operational status of a water storage facility can include closed, water storage, water discharge, and simultaneous inflow and discharge. For example, if the inlet gate is closed and the outlet gate is open, the facility opening and closing information is "Inlet gate closed, outlet gate open." Correspondingly, the operational status of the water storage facility at this time is water discharge. As another example, if both the inlet and outlet gates are open, the facility opening and closing information is "Inlet gate open, outlet gate open." Correspondingly, the operational status of the water storage facility at this time is simultaneous inflow and discharge.

[0092] In some embodiments, in response to the water accumulation risk meeting a first preset condition, the management platform can determine a first joint control scheme corresponding to the target area based on regional information.

[0093] The first preset condition can refer to the condition that the water accumulation risk of all roads or intersections in the target area meets. For example, the first preset condition can be that the sum of the water accumulation risks of all roads or intersections in the target area exceeds a first water accumulation risk threshold. The first water accumulation risk threshold can be preset manually.

[0094] The first joint regulation scheme can refer to a scheme that controls the opening and closing status of storage facilities in the target area, thereby storing or diverting rainwater and floods in the target area.

[0095] In some embodiments, the first joint control scheme includes a joint scheduling scheme for storage facilities in the target area. The joint control scheme for storage facilities can refer to a scheme that controls the opening and closing status of all storage facilities in the target area. The operating status of the storage facilities can include their opening and closing status, current drainage flow, etc. For example, the content of the first joint control scheme could be: "Storage facility 1 is in drainage state, and the drainage flow rate is set to 8m³ / s." 3 / h; Storage facility 2 is in the closed state; Storage facility 3 is in the drainage state, and the drainage flow rate is set to 10m³. 3 / h......”.

[0096] In some embodiments, the management platform can use the historical control scheme corresponding to the historical first information that is most similar to the current first information as the first joint control scheme to be used.

[0097] Figure 4-1 This is an exemplary structural diagram of a joint scheduling model shown in some embodiments of this specification. In some embodiments, the joint scheduling model 430 may be executed by a management platform.

[0098] In some embodiments, the management platform can determine a first joint control scheme corresponding to the target area based on the processing of regional information using a joint scheduling model. The joint scheduling model is a machine learning model.

[0099] In some embodiments, the input to the joint scheduling model may be regional information of the target area, and the output may be the opening and closing status of each storage facility in the first joint control scheme.

[0100] In some embodiments, the joint scheduling model can be trained using multiple labeled training samples. For example, the management platform can input multiple labeled training samples into an initial joint scheduling model, construct a loss function using the labels and the results of the initial joint scheduling model, and iteratively update the parameters of the initial joint scheduling model based on the loss function. The model training is complete when the loss function of the initial joint scheduling model satisfies a preset condition, resulting in a trained joint scheduling model. The preset condition could be loss function convergence, the number of iterations reaching a threshold, etc.

[0101] In some embodiments, the training samples may be historical regional information of the historical target area, and the labels may be historical joint control schemes corresponding to the historical regional information.

[0102] In some embodiments, the management platform can determine the regional road network infrastructure map structure data of the target area based on the regional information of the target area, process the regional road network infrastructure map structure data based on the joint scheduling model, and determine the first joint control scheme corresponding to the target area.

[0103] In some embodiments, such as Figure 4-1 As shown, the joint scheduling model 430 can determine the first joint control scheme 440 based on the regional road network infrastructure map structure data 420 of the target area determined by the regional information 410 of the target area.

[0104] Figure 4-2 This is a schematic diagram of regional road network infrastructure map structure data according to some embodiments of this specification. In some embodiments, the regional road network infrastructure map structure data 420 may be generated by a management platform based on regional information of a target area.

[0105] In some embodiments, the regional road network infrastructure map structure data is a data structure composed of nodes and edges, with edges connecting nodes, and nodes and edges may have attributes.

[0106] In some embodiments, the nodes in the regional road network infrastructure map structure data may include a first node and a second node. The first node may be an intersection node, and the second node may be a water storage facility node. For example, as... Figure 4-2 As shown, the regional road network infrastructure map structure data 420 may include first node 1, first node 2, first node 3, first node 4, first node 5, second node 1, and second node 2.

[0107] In some embodiments, the first node may correspond to an intersection within the target area. The attributes of the first node may reflect the relevant characteristics of the intersection. For example, the attributes of the first node may include rainfall information, drainage capacity information, and historical water accumulation information of the intersection.

[0108] In some embodiments, the second node may correspond to a water storage facility within the target area. The attributes of the second node may reflect the relevant characteristics of the water storage facility. For example, the attributes of the second node may include the location information, water storage capacity information, drainage capacity information, and opening / closing information of the water storage facility.

[0109] In some embodiments, the edges of the device graph structure data may include a first edge, a second edge, and a third edge. The first edge is a one-way edge pointing from a first node with a higher elevation to a first node with a lower elevation. Further, the first edge may correspond to a road between intersections corresponding to the first nodes.

[0110] The second edge can connect the first node and the second node based on the direct connection of drainage pipes between the intersection corresponding to the first node and the storage facility corresponding to the second node. The second edge is a one-way edge from the first node to the second node. Furthermore, the second edge can correspond to the drainage pipe set between the intersection corresponding to the first node and the storage facility corresponding to the second node.

[0111] The third edge can connect to the second node based on the water flow direction between the storage facilities corresponding to the second node. The third edge is a one-way edge from the second node corresponding to the storage facility through which the water flows first to the second node corresponding to the storage facility through which the water flows later. Furthermore, the third edge can correspond to the drainage pipes, ditches, or rivers between the storage facilities corresponding to the second node.

[0112] For example, such as Figure 4-2 As shown, the regional road network infrastructure map structure data 420 may include first side a, first side b, first side c, first side d, first side e, first side f, second side a, second side b, and third side a.

[0113] In some embodiments, the attributes of the first side may include precipitation information, drainage capacity information, historical water accumulation information, slope information, and length information of the corresponding road. The slope information refers to the degree of inclination of the corresponding road, which can be characterized by the inclination angle. The length information refers to the length of the corresponding road. The attributes of the second side may include drainage capacity information and length information of the corresponding pipeline. The attributes of the third side may include drainage capacity information and length information of the corresponding pipeline. If the third side corresponds to a ditch or river, the attributes of the third side may also include information such as the water flow velocity, water surface width, and water depth of the ditch or river.

[0114] In some embodiments, the joint scheduling model can be a graph neural network (GNN) model. The input of the joint scheduling model can be regional road network infrastructure map structure data, and the output can be the first joint control scheme corresponding to the target area, wherein the node outputs in the GNN correspond to the on / off status information of the storage and regulation equipment.

[0115] The joint scheduling model can be trained based on training data. Training data includes training samples and labels. For example, training samples can be historical graphs determined based on historical data. The nodes and their attributes, edges and their attributes of the historical graphs are similar to the structure data of the regional road network infrastructure graph. Labels can be the on / off status information of historical water storage facilities.

[0116] The regional road network infrastructure map structure data described in some embodiments of this specification is used as input to the joint scheduling model, so that the opening and closing status of each storage and regulation device is planned with reference to information on relevant intersections, roads, storage and regulation devices, etc., making the opening and closing status of each storage and regulation device in the first joint control scheme more adaptable to the requirements of drainage operations in the target area.

[0117] The joint scheduling model described in some embodiments of this specification analyzes and obtains the first joint control scheme for the target area by taking the regional information of the target area as input, so that the planned scheme is more adapted to the requirements of drainage operations in the target area.

[0118] The method for determining the first joint control scheme of the target area based on regional information, as described in some embodiments of this specification, effectively reduces the manpower and time costs of manually planning control schemes, while enhancing the adaptability of the scheme to actual drainage needs.

[0119] In some embodiments, the area information may further include third and fourth information.

[0120] The third piece of information may include the division of demand points in the target area, demand information at each demand point, water accumulation risk at each demand point, and priority coefficient for each demand point. Demand points can be roads or intersections within the target area that meet the second preset condition. The second preset condition refers to the fact that the water accumulation risk of at least one road segment or intersection within the target area still exceeds a second threshold after being regulated by the first joint control scheme. The second threshold can be manually preset.

[0121] In some embodiments, water accumulation risk may refer to the water accumulation risk of each road or intersection re-acquired after the implementation of the first joint control scheme. The method for re-acquiring water accumulation risk may be related to... Figure 2 The method described in the text is the same.

[0122] The information on demand point classification can include the demand point's number, location information, and the correspondence between the demand point and the rescue point. The correspondence means that when a demand point needs rescue supplies, the corresponding rescue point will provide them. Details regarding rescue points will be provided later. For example, the classification information for a demand point could be: "Demand point 1, located at the intersection of [road name] and [road name], corresponding rescue point is rescue point 4."

[0123] The demand information for a demand point can include the demand point number, the types of relief supplies needed, and the quantity of each type of relief supplies required. For example, the demand information for a demand point could be "Demand point 1, 20kg of emergency food, 10 life jackets".

[0124] The risk of water accumulation at demand points can refer to the risk of water accumulation on the roads or intersections corresponding to the demand points.

[0125] A demand point priority coefficient refers to the degree of priority a demand point receives relief supplies from a relief point. A higher priority coefficient indicates a higher priority for the demand point to receive relief supplies. For example, the demand point priority coefficient could be "Demand Point 1, Priority Coefficient 0.3", "Demand Point 2, Priority Coefficient 0.8", "Demand Point 3, Priority Coefficient 0.1", etc. In some embodiments, the priority coefficient can be determined based on the type of public place near the demand point: each public place corresponds to a priority coefficient value, and the priority coefficient of the demand point is the sum of the priority coefficient values ​​of all nearby public places. The priority coefficient value corresponding to each public place can be preset, for example, 0.4 for hospitals, 0.5 for schools, 0.2 for shopping malls, etc.

[0126] The fourth piece of information may include the division of rescue points in the target area, the information on supplies at the rescue points, and the historical rescue information of the rescue points. A rescue point may refer to a location within the target area that stores rescue supplies and can supply those supplies to the points of need when necessary.

[0127] The information regarding the division of rescue points can include the rescue point's number, location information, and the correspondence between the demand points and the rescue points. The correspondence means that when a demand point needs rescue supplies, the corresponding rescue point will provide them. For example, the information regarding the division of a rescue point could be "Rescue Point 1, located at No. 1, [Road Name], corresponding demand points include Demand Point 2 and Demand Point 5".

[0128] Information on supplies at a rescue point can include the types of supplies stored there, as well as the quantity of each type. For example, information on supplies at a rescue point could include "100kg of emergency food, 200 life jackets, 5 lifeboats, and 5 oxygen cylinders".

[0129] Historical rescue information at a rescue point can point to historical records of the types and quantities of rescue supplies supplied to the requesting points. For example, historical rescue information for a certain rescue point could be: "On a certain date, 10 kg of emergency food was supplied to requesting point 1; on a certain date, 20 kg of emergency food was supplied to requesting point 3; and on a certain date, 30 life jackets were supplied to requesting point 2."

[0130] In some embodiments, the management platform may, in response to the fact that the risk of water accumulation at least one road segment or intersection within the target area meets a second preset condition, determine the material dispatch plan corresponding to the target area based on the area information.

[0131] A material dispatch plan refers to the arrangement of the types and quantities of relief supplies supplied from rescue points within a target area to points in need. For example, the material dispatch plan could include: "Rescue point 1 supplies 10kg of emergency food and 20 life jackets to point 1 in need, and supplies 3 lifeboats to rescue point 2; rescue point 2 supplies 20 life jackets to point 1 in need, and supplies 20kg of emergency food to rescue point 3; rescue point 3 supplies 30 life jackets, 15kg of emergency food, and 3 lifeboats to point 4 in need, and supplies 5kg of emergency food to rescue point 1..."

[0132] In some embodiments, the management platform can determine the corresponding material dispatching plan for the target area based on the area information of the target area using a machine learning model. The input to the machine learning model can be the area information of the target area, and the output can be the corresponding material dispatching plan for the target area.

[0133] Figure 5 This is an exemplary flowchart for determining the material dispatching plan corresponding to the target area. For example... Figure 5 As shown, the process 500 for determining the material dispatch plan corresponding to the target area may include the following steps. Process 500 can be executed by the management platform.

[0134] Step 510: The management platform can obtain the initial material dispatch plan. The initial material dispatch plan can include the initial supply quantities of materials from each rescue point to each demand point. There can be more than one initial material dispatch plan. The content of the initial material dispatch plan can be system default values ​​set according to actual needs, experience values, manually preset values, or any combination thereof.

[0135] In some embodiments, the initial material scheduling scheme can be represented by an initialization vector. An exemplary process for initializing the vector corresponding to the initial material scheduling scheme is as follows: For a number of demand points (let the quantity be J, where J is a positive integer) and a number of rescue points (let the quantity be I, where I is a positive integer), the management platform can set the number of initial material scheduling schemes to N (N is a positive integer), and the vector corresponding to the nth initial material scheduling scheme... It can be represented as:

[0136]

[0137] The vector corresponding to N initial candidate irrigation schemes It can be represented as:

[0138]

[0139] Where 0 is an identifier (representing the 0th iteration, i.e., the initial value before iteration begins); n (n is a positive integer) represents the number of the initial material scheduling scheme, and n≤N; for example, The quantity of the m-th type of relief supplies supplied by relief point i to demand point j, where i, j, and m are all positive integers, and i ≤ I, j ≤ j, m ≤ M, where M refers to the total number of types of relief supplies stored at all relief points.

[0140] In some embodiments, during the initialization phase of the vector corresponding to the initial material scheduling scheme, let the total amount of the k-th type of rescue material stored at rescue point i be U. im The quantity of the m-th type of relief supplies supplied from relief point i to demand point 1 is randomly initialized to... Then the m-th type of relief supplies supplied from relief point i to demand point 2 should be based on the remaining quantity of the m-th type of relief supplies at relief point i. Randomly initialize for the upper boundary And so on.

[0141] Step 520: The management platform can iteratively update the initial material scheduling scheme based on the first preset algorithm to obtain at least one candidate material scheduling scheme.

[0142] In some embodiments, the iterative update steps of the first preset algorithm may include: obtaining the adjustment range to be updated; updating the adjustment range to obtain the updated adjustment range; and updating the initial material scheduling scheme based on the updated adjustment range to obtain a candidate material scheduling scheme. The adjustment range to be updated in the first iteration can be preset. Furthermore, the management platform can use the candidate material scheduling scheme as the initial material scheduling scheme for the next iteration, and the updated adjustment range as the adjustment range to be updated in the next iteration.

[0143] In some embodiments, the adjustment range to be updated may include the adjustment range of the supply of each type of material from each rescue point to each demand point in the candidate material dispatch scheme.

[0144] In each subsequent iteration, the adjustment range to be updated is updated to obtain the updated adjustment range. Based on the updated adjustment range, the candidate material scheduling scheme to be processed is updated to obtain the updated candidate material scheduling scheme. The updated candidate material scheduling scheme is determined as the initial material scheduling scheme for the next round, and the updated adjustment range is determined as the adjustment range to be updated for the next round.

[0145] In some embodiments, the adjustment range to be updated can be data in vector form, where each element of each dimension can be considered an incremental element to be updated. Updating the adjustment range can be achieved by updating the incremental elements to be updated. The adjustment range to be updated can contain multiple incremental elements to be updated. There can be a one-to-one correspondence between each element and each incremental element in the vector corresponding to the candidate material scheduling scheme to be processed. An incremental element can be used to characterize the adjustment range of the supply of a certain material from a rescue point to a demand point in the candidate material scheduling scheme.

[0146] In some embodiments, the initial value of the adjustment range to be updated can be represented by an initialization vector. For example, the vector corresponding to the initial adjustment range to be updated for the nth initial material scheduling scheme. It can be represented as:

[0147]

[0148] This can be used in the first iteration to iteratively update the nth initial material scheduling scheme. For example, the vector corresponding to the initial adjustment magnitude to be updated for each of the N initial material scheduling schemes. It can be represented as:

[0149]

[0150] In some embodiments, the management platform can update the incremental elements to be processed based on the current loss of the previous round, and the updated incremental elements will serve as the incremental elements to be updated in the next round. The current loss of the previous round can be determined based on the difference in effectiveness between the candidate material scheduling scheme obtained in the previous round and the historically optimal candidate material scheduling scheme. The difference in effectiveness can be determined based on the evaluation value of the candidate material scheduling scheme. The definition and calculation method of the evaluation value are described in the relevant section below.

[0151] For example, after the (k+1)th iteration, the updated incremental element can be calculated using the following formula:

[0152]

[0153] Where n represents the number of the candidate material scheduling scheme. k represents the number of iteration rounds, where k≥0. This represents the incremental element to be processed corresponding to the nth candidate material scheduling scheme in the (k+1)th iteration. This represents the nth candidate resource scheduling scheme obtained after the kth iteration. ω represents the inertia weight constant. c1 represents the individual learning factor, and c2 represents the group learning factor. r1 and r2 are arbitrary values ​​in the interval [0,1], used to increase the randomness of the search.

[0154] Let be the optimal solution of the nth candidate material scheduling scheme after the kth iteration, across all iterations. The optimal solution here can refer to the material scheduling scheme (i.e., the individual historical optimal solution) where the evaluation value of the nth candidate material scheduling scheme after the kth iteration is the minimum among the evaluation values ​​calculated after each iteration.

[0155] This refers to the optimal solution among all N candidate material scheduling schemes after the k-th iteration, considering all iterations. The optimal solution here can be the material scheduling scheme with the smallest evaluation value among the optimal solutions of the N candidate material scheduling schemes after the k-th iteration (i.e., the group's historical optimal solution). The aforementioned inertia weight constant, individual learning factor, group learning factor, and random constant can be system default values, empirical values, pre-set values, or any combination thereof, set according to actual needs.

[0156] In some embodiments, each candidate material scheduling scheme can be updated based on the incremental element in the updated adjustment range. For example, after the (k+1)th iteration, the updated candidate material scheduling scheme can be calculated using the following formula:

[0157]

[0158] For example, after the first iteration, the updated nth candidate irrigation scheme can be calculated using the following formula:

[0159]

[0160] In some embodiments, the adjustment range to be updated and the updated adjustment range need to ensure that the candidate material scheduling scheme updated based on the adjustment range meets preset constraints.

[0161] In some embodiments, the constraints include the following three relationships:

[0162]

[0163]

[0164]

[0165] Where, q jm The demand for the m-th type of relief supplies represents the demand at demand point j; Equation (1) indicates that the quantity of various types of relief supplies supplied to each demand point does not exceed its demand; c imEquation (2) represents the total amount of the m-th type of relief supplies stored at the i-th relief point; Equation (2) indicates that the amount of relief supplies sent out by the relief point is equal to its available supply, that is, in the case of supply falling short of demand, all relief supplies should be distributed; Z represents an integer; Equation (3) indicates that the amount of relief supplies supplied by the relief point to each demand point is a non-negative integer.

[0166] In some embodiments, if the updated adjustment range does not meet preset constraints, the adjustment range value that is closest to the actually calculated updated adjustment range among the adjustment range values ​​that meet the preset constraints is taken as the updated adjustment range and used in the next iteration. For example, in a certain updated adjustment range, there exists a ∑ i∈I ∑ j∈J ∑ m∈M X ijm >c im Then X i1m To X iJm The values ​​of each item are determined by (c im / ∑ i∈I ∑ j∈J ∑ m∈M X ikm After the proportion of ) is reduced, it will be included in the candidate material scheduling plan.

[0167] Step 530: The management platform can calculate the evaluation value of each candidate material scheduling scheme in at least one candidate material scheduling scheme based on the second preset algorithm.

[0168] The second preset algorithm refers to the algorithm used to calculate the evaluation value of candidate material scheduling schemes.

[0169] In some embodiments, the second preset algorithm may include two objective functions. For example, the evaluation value may be a weighted sum of the results (f1 and f2) of the two objective functions. In some embodiments, the two objective functions are as follows:

[0170]

[0171]

[0172] Where max represents the maximum value; r j The risk of water accumulation at demand point j; μ j The priority coefficient representing demand point j; d ij This represents the number of times relief supplies are transported from relief point i to need point j; s ij This represents the distance traveled by the means of transport used when delivering relief supplies from relief point i to the point j in need. In equation (4), (∑ i∈I ∑ m∈M X ijm) / (∑ m∈M q jm Let represent the degree to which each demand point j among J demand points is satisfied. The degree of satisfaction can refer to the proportion of relief supplies received by a demand point relative to the total demand for supplies. For example, demand point j receives X quantities of the first, second, and third types of relief supplies respectively. j1 X j2 X j2 The demand for the first, second, and third types of relief supplies is q respectively. j1 q j2 q j3 Then the degree of satisfaction of demand point j = (X j1 +X j2 +X j2 ) / (q j1 +q j2 +q j3 Corresponding to (1-(∑) i∈I ∑ m∈M X ijm ) / (∑ m∈M q jm )) represents the degree of dissatisfaction of each requirement point j out of J requirements, (r j μ j (1-(∑ i∈I ∑ m∈M X ik, ) / (∑ m∈M q jm ))) represents the degree of non-compliance of the demand point water accumulation risk and priority coefficient. In equation (5), ∑ i∈I ∑ j∈J d ij s ij This can represent the total transportation cost of transporting relief supplies from various relief points to various points of need.

[0173] The evaluation value can be calculated based on the following formula:

[0174] p = h1f1 + h2f2

[0175] Where p represents the evaluation value; h1 and h2 represent the weight values ​​of f1 and f2, respectively. The values ​​of h1 and h2 can be preset manually.

[0176] In some embodiments, the evaluation value of candidate material dispatching schemes may be related to the road flooding risk in the target area. In some embodiments, the road flooding risk is determined based on a flooding risk prediction model. The flooding risk prediction model is a machine learning model, and its structure can be related to... Figure 2 The flooding risk prediction model is the same. See details below. Figure 2 The corresponding content.

[0177] Step 540: The management platform can iterate the candidate material scheduling plan multiple times until the iteration completion conditions are met.

[0178] In some embodiments, the iteration completion condition can be that when more than one candidate material scheduling scheme converges into one candidate material scheduling scheme, its evaluation value is less than an evaluation value threshold. The evaluation value threshold can be preset manually.

[0179] In some other embodiments, the iteration completion condition may be that the number of iterations reaches a preset number.

[0180] Ultimately, the management platform can use the candidate material scheduling scheme obtained when the iteration completion conditions are met as the final material scheduling scheme.

[0181] Some embodiments in this specification can iteratively update and optimize multiple candidate material dispatch schemes, thereby determining the final material dispatch scheme that meets the evaluation requirements. This saves transportation costs while maximizing the availability of relief supplies to meet the material needs of the target area.

[0182] Some embodiments in this specification, by introducing a comprehensive analysis of the material demand roads or intersections and relief material storage points in the target area, maximize the satisfaction of the material demand points in the target area.

[0183] It should be noted that the above description of process 500 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 500 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification. For example, adjusting the weight values ​​used in calculating the evaluation values.

[0184] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0185] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0186] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0187] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0188] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0189] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0190] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and are considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for predicting waterlogging risk in smart cities, implemented based on the management platform of a smart city waterlogging risk prediction IoT system, the method comprising: Based on the acquired regional information of the target area, the flood risk of the target area is predicted. The regional information includes third and fourth information of the target area: the third information includes the division information of demand points in the target area, demand point demand information, flood risk of demand points, and demand point priority coefficient; the fourth information includes the division information of rescue points in the target area, rescue point material information, and historical rescue information of rescue points. Based on the aforementioned waterlogging risk, a corresponding adjustment plan for the target area is determined, including: In response to the flooding risk meeting a second preset condition, based on the regional information, a material dispatching plan corresponding to the target region is determined, including: Obtain an initial material dispatch plan, which includes the initial supply amount of rescue materials from each of the rescue points to each of the demand points; The initial material scheduling scheme is iteratively updated based on the first preset algorithm to obtain at least one candidate material scheduling scheme. An evaluation value is calculated for each candidate material scheduling scheme in the at least one candidate material scheduling scheme based on a second preset algorithm. The evaluation value is determined based on the result values ​​of two objective functions. The method for determining the evaluation value includes: ,in, The evaluation value is... f 1 and f 2 represents the result values ​​of the two objective functions, respectively. and These are the weights of the result values ​​of the two objective functions, respectively; result values The result value is determined based on the water accumulation risk at the demand point, the priority coefficient of the demand point, and the degree to which the demand point is satisfied. The methods for determining this include: ,in, The number of rescue points in the target area. The number of required points in the target area. This refers to the type and quantity of the relief supplies. To find the sign of the maximum value, For demand points The aforementioned demand points pose a risk of water accumulation. For the aforementioned demand point The priority coefficient of the aforementioned demand points, For the aforementioned demand point Degree of satisfaction; Result value f 2. The result value is determined based on the number of times the relief supplies are delivered and the distance traveled by the transport vehicles. The methods for determining this include: ,in, For the rescue point To the aforementioned demand point The number of times the relief supplies were transported. The rescue point To the aforementioned demand point The distance traveled by the transport vehicle when delivering the relief supplies; The at least one candidate material scheduling scheme is iterated multiple times until the iteration completion condition is met. The iteration completion condition includes that when multiple candidate material scheduling schemes converge into one candidate material scheduling scheme, the evaluation value of the candidate material scheduling scheme is less than the evaluation value threshold. Execute the adjustment instructions corresponding to the adjustment scheme.

2. The method according to claim 1, wherein the regional information further includes: road network information, precipitation information, drainage capacity information, and historical water accumulation information of the target region; The prediction of water accumulation risk in the target area based on the acquired regional information includes: Based on the processing of regional information using a water accumulation risk prediction model, the road water accumulation risk of each road in the target region during the target time period is determined. The water accumulation risk prediction model is a machine learning model.

3. The method according to claim 1, wherein the regional information further includes first information and second information of the target region: the first information includes road network information, precipitation information, drainage capacity information, and historical water accumulation information of the target region; the second information includes the deployment information of water storage facilities, water storage capacity information of the facilities, drainage capacity information of the facilities, and facility opening and closing information of the target region; The method for determining the adjustment plan corresponding to the target area based on the water accumulation risk further includes: In response to the water accumulation risk meeting a first preset condition, a first joint control scheme corresponding to the target area is determined based on the regional information; The first joint control scheme includes a joint scheduling scheme for the storage facilities in the target area.

4. The method according to claim 3, wherein determining the first joint control scheme corresponding to the target area based on the regional information includes: Based on the processing of the regional information using the joint scheduling model, a first joint control scheme corresponding to the target region is determined. The joint scheduling model is a machine learning model.

5. The method according to claim 1, wherein the IoT system for predicting waterlogging risk in smart cities further comprises: User platform, service platform, sensor network platform, object platform; The management platform includes a central database and several sub-platforms. The sensor network platform includes several sensor network sub-platforms; the regional information is obtained based on the object platform and transmitted by the corresponding sensor network sub-platform to the corresponding management sub-platform, and then uploaded by the management sub-platform to the overall database; The method further includes: The adjustment plan is transmitted to the user platform based on the service platform.

6. A smart city flood risk prediction IoT system, characterized in that, The system includes: a management platform, a user platform, a service platform, a sensor network platform, and an object platform. The management platform includes a main database and several management sub-platforms. The sensor network platform includes several sensor network sub-platforms. Regional information is obtained based on the object platform and transmitted from the corresponding sensor network sub-platform to the corresponding management sub-platform, and then uploaded to the overall database by the management sub-platform. The management platform is configured to perform the following operations: Based on the acquired regional information of the target area, the flood risk of the target area is predicted. The regional information includes third and fourth information of the target area: the third information includes the division information of demand points in the target area, demand point demand information, flood risk of demand points, and demand point priority coefficient; the fourth information includes the division information of rescue points in the target area, rescue point material information, and historical rescue information of rescue points. Based on the aforementioned waterlogging risk, a corresponding adjustment plan for the target area is determined, including: In response to the flooding risk meeting a second preset condition, based on the regional information, a material dispatching plan corresponding to the target region is determined, including: Obtain an initial material dispatch plan, which includes the initial supply amount of rescue materials from each of the rescue points to each of the demand points; The initial material scheduling scheme is iteratively updated based on the first preset algorithm to obtain at least one candidate material scheduling scheme. An evaluation value is calculated for each candidate material scheduling scheme in the at least one candidate material scheduling scheme based on a second preset algorithm. The evaluation value is determined based on the result values ​​of two objective functions. The method for determining the evaluation value includes: ,in, The evaluation value is... f 1 and f 2 represents the result values ​​of the two objective functions, respectively. and These are the weights of the result values ​​of the two objective functions, respectively; result values The result value is determined based on the water accumulation risk at the demand point, the priority coefficient of the demand point, and the degree to which the demand point is satisfied. The methods for determining this include: ,in, The number of rescue points in the target area. The number of required points in the target area. This refers to the type and quantity of the relief supplies. To find the sign of the maximum value, For demand points The aforementioned demand points pose a risk of water accumulation. For the aforementioned demand point The priority coefficient of the aforementioned demand points, For the aforementioned demand point Degree of satisfaction; Result value f 2. The result value is determined based on the number of times the relief supplies are delivered and the distance traveled by the transport vehicles. The methods for determining this include: ,in, For the rescue point To the aforementioned demand point The number of times the relief supplies were transported. The rescue point To the aforementioned demand point The distance traveled by the transport vehicle when delivering the relief supplies; The at least one candidate material scheduling scheme is iterated multiple times until the iteration completion condition is met. The iteration completion condition includes that when multiple candidate material scheduling schemes converge into one candidate material scheduling scheme, the evaluation value of the candidate material scheduling scheme is less than the evaluation value threshold. Execute the adjustment instructions corresponding to the adjustment scheme.

7. A smart city waterlogging risk prediction device, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing computer instructions that, when executed by a processor, implement the smart city flood risk prediction method as described in any one of claims 1 to 5.

Images