Evacuation path optimization method and device in flood scene and electronic equipment

Abstract

The invention provides an evacuation path optimization method and device in a flood scene and electronic equipment. The method comprises the steps of obtaining building space parameters of a target building; performing hydrodynamic simulation on a preset flood scene based on the building space parameters to obtain flood data; collecting spatial position distribution information of to-be-evacuated objects in the space of the target building; constructing a point queue prediction model based on the spatial position distribution information and the to-be-evacuated objects; based on the flood data, the spatial position distribution information, the point queue prediction model and a preset MPC model prediction control framework, constructing a prediction control optimization model for the target building in the flood scene; an evacuation prediction control optimization result of the target building in the flood scene is obtained through the prediction control optimization model and an interior point method-based mixed solution framework optimized by using a speed pause particle swarm algorithm; and evacuation indication information is sent to the to-be-evacuated object based on the evacuation prediction control optimization result.

Description

Technical Field

[0001] This disclosure relates to the field of emergency safety technology, specifically to a method, apparatus, and electronic equipment for optimizing evacuation routes in flood scenarios. Background Technology

[0002] With the rapid pace of urbanization, complex building structures such as high-rise buildings, large commercial complexes, and underground transportation hubs are becoming increasingly common, leading to a continuous increase in the density of people inside these buildings. Simultaneously, due to intensified global climate change and frequent extreme weather events, urban flooding and other water-related disasters pose a serious threat to building safety. In flood scenarios, the safe evacuation of people from buildings has become a critical technical issue that urgently needs to be addressed.

[0003] Traditional evacuation methods rely primarily on static evacuation plans and manual command, lacking the ability to dynamically respond to real-time environmental changes. Existing technologies often use pre-set, fixed routes for evacuation path planning, failing to adequately consider the impact of dynamic environmental factors such as rising water levels, blocked passageways, and strong water currents during floods. This results in low evacuation efficiency and may even lead people to more dangerous areas. Summary of the Invention

[0004] This invention proposes a method, device, and electronic equipment for optimizing evacuation routes in flood scenarios.

[0005] In a first aspect, embodiments of the present invention propose a method for optimizing evacuation routes in a flood scenario, comprising: acquiring architectural spatial parameters of a target building; performing hydrodynamic simulation on a preset flood scenario based on the architectural spatial parameters to obtain flood data; collecting spatial location distribution information of objects to be evacuated within the space of the target building; constructing a prediction model based on a point queue network based on the spatial location distribution information and the objects to be evacuated; constructing a prediction and control optimization model for the target building in the flood scenario based on the flood data, spatial location distribution information, point queue prediction model, and a preset MPC model prediction and control framework; obtaining the evacuation prediction and control optimization result of the target building in the flood scenario through the prediction and control optimization model and a hybrid solution framework based on the interior point method optimized using the velocity-pause particle swarm optimization algorithm; and sending evacuation instruction information to the objects to be evacuated based on the evacuation prediction and control optimization result.

[0006] Secondly, embodiments of the present invention propose an evacuation route planning device for flood scenarios, comprising: a building parameter acquisition module configured to acquire building spatial parameters of a target building; a flood parameter acquisition module configured to perform hydrodynamic simulation of a preset flood scenario based on the building spatial parameters to obtain flood data; a spatial distribution information acquisition module configured to acquire spatial location distribution information of objects to be evacuated within the space of the target building; a point queue prediction model construction module configured to construct a point queue prediction model based on the spatial location distribution information and the objects to be evacuated; a prediction control optimization model construction module configured to construct a prediction control optimization model for the target building in the flood scenario based on the flood data, spatial location distribution information, point queue prediction model, and a preset MPC model prediction control framework; an evacuation prediction control optimization result module configured to obtain the evacuation prediction control optimization result of the target building in the flood scenario through the prediction control optimization model and a hybrid solution framework based on the interior point method optimized using the velocity pause particle swarm algorithm; and an evacuation instruction sending module configured to send evacuation instruction information to the objects to be evacuated based on the evacuation prediction control optimization result.

[0007] Thirdly, embodiments of the present invention provide an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to implement the evacuation route optimization method in a flood scenario as described in any implementation of the first aspect.

[0008] Fourthly, embodiments of the present invention provide a non-transitory computer-readable storage medium storing computer instructions that enable a computer to implement an evacuation route optimization method in a flood scenario as described in any of the implementations in the first aspect.

[0009] Fifthly, embodiments of the present invention provide a computer program product including a computer program, which, when executed by a processor, can implement the evacuation route optimization method in a flood scenario as described in any of the implementations in the first aspect.

[0010] This invention provides a method, device, and electronic equipment for optimizing evacuation routes in flood scenarios. Based on flood dynamic parameters and the dynamics of personnel evacuation, a predictive control optimization model is established. The control results are obtained based on a gradient-information-based hybrid solution framework optimized using a velocity-pause particle swarm optimization algorithm. This ensures the stability and reliability of the crowd evacuation route control results and improves the safety and effectiveness of personnel evacuation from buildings in flood scenarios.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0013] Figure 1 This is an exemplary system architecture to which this disclosure can be applied;

[0014] Figure 2 A flowchart of an evacuation route optimization method in a flood scenario provided by an embodiment of the present invention;

[0015] Figure 3 This is a schematic diagram of a pedestrian evacuation simulation scenario inside a target building provided in an embodiment of the present invention;

[0016] Figure 4 A schematic diagram of the PQ network corresponding to the pedestrian evacuation simulation scenario inside the target building provided in an embodiment of the present invention;

[0017] Figure 5 A schematic diagram illustrating the details of the number of people transferred between exits within 360 seconds in a pedestrian evacuation simulation scenario within a target building on the third basement level of a platform, as provided in an embodiment of the invention.

[0018] Figure 6 and Figure 7 A schematic diagram of the number of people evacuated from platform B3 within 360 seconds under different evacuation strategies in a pedestrian evacuation simulation scenario within a target building, provided for an embodiment of the invention.

[0019] Figure 8 A flowchart of another evacuation route optimization method in a flood scenario provided by an embodiment of the present invention;

[0020] Figure 9 A structural block diagram of an evacuation route planning device for a flood disaster scenario provided in an embodiment of the present invention;

[0021] Figure 10 This is a schematic diagram of the structure of an electronic device suitable for performing an evacuation route optimization method in a flood scenario, as provided in an embodiment of the present invention. Detailed Implementation

[0022] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments of the invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0023] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0024] Figure 1 An exemplary system architecture 100 is shown, in which embodiments of the evacuation route optimization methods, apparatus, electronic devices, and computer-readable storage media in flood scenarios disclosed herein can be applied.

[0025] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0026] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various applications for enabling information communication between the terminal devices 101, 102, and 103 and server 105 can be installed. These applications include instant messaging applications.

[0027] Terminal devices 101, 102, and 103 and server 105 can be either hardware or software. When terminal devices 101, 102, and 103 are hardware, they can be various electronic devices with displays, including but not limited to smartphones, tablets, laptops, and desktop computers. When terminal devices 101, 102, and 103 are software, they can be installed in the aforementioned electronic devices, and can be implemented as multiple software programs or software modules, or as a single software program or software module; no specific limitation is made here. When server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When server 105 is software, it can be implemented as multiple software programs or software modules, or as a single software program or software module; no specific limitation is made here.

[0028] Server 105 can provide various services through its built-in applications. It should be noted that the data or information required to provide these services can be obtained from terminal devices 101, 102, and 103 via network 104, or it can be pre-stored locally on server 105 through various means. Therefore, when server 105 detects that this data is already stored locally, it can choose to retrieve it directly from the local storage. In this case, the exemplary system architecture 100 may not include terminal devices 101, 102, and 103 and network 104.

[0029] Since providing various services may require significant computing resources and strong computing power, the evacuation route optimization methods for flood scenarios provided in the subsequent embodiments of this disclosure are generally executed by a server 105 with strong computing power and abundant computing resources. Correspondingly, the evacuation route planning device for flood scenarios is also generally located in the server 105. However, it should also be noted that when terminal devices 101, 102, and 103 also possess sufficient computing power and resources, they can also complete the aforementioned calculations performed by the server 105 through their installed applications, thereby outputting the same results as the server 105. Especially when multiple terminal devices with different computing capabilities exist simultaneously, but the relevant application determines that the terminal device has strong computing power and abundant remaining computing resources, the terminal device can perform the aforementioned calculations, thereby appropriately reducing the computing pressure on the server 105. Accordingly, the evacuation route planning device for flood scenarios can also be located in the terminal devices 101, 102, and 103. In this case, the exemplary system architecture 100 may also exclude the server 105 and the network 104.

[0030] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0031] Please refer to Figure 2 , Figure 2 A flowchart of an evacuation route optimization method in a flood scenario provided by an embodiment of the present invention, wherein process 200 includes the following steps:

[0032] Step 201: Obtain the architectural space parameters of the target building.

[0033] In this embodiment, the evacuation route optimization method in a flood scenario is implemented by the entity (e.g., Figure 1 The server 105 shown obtains the building space parameters of the target building. This target building is considered a potential site of flooding, such as... Figure 3As shown, its architectural space parameters may include the structural data of the target building (such as key areas such as rooms and staircases).

[0034] Step 202: Perform hydrodynamic simulation on the preset flood scenario based on building space parameters to obtain flood data.

[0035] In this embodiment, the aforementioned execution entity performs a hydrodynamic simulation of a preset flood scenario based on the building space parameters to obtain flood data of the target building under the simulated flood scenario. For example, the building space parameters mainly include the structural data of the target building. Based on the structural data and the preset hydrodynamic model, a three-dimensional simulation model of the target building in the flood scenario is established, and a preset flood scenario simulation is performed for possible indoor flooding. Monitoring points are set up in the simulation model inside the building to collect flood data throughout the entire flood event. All monitored flood data includes water depth, flow velocity, and inlet flow rate, etc.

[0036] Step 203: Collect spatial location distribution information of the objects to be evacuated within the target building.

[0037] In this embodiment, the evacuees mainly refer to the crowds within the target building. Spatial distribution information of the evacuees within the current building space can be collected by setting up pedestrian counting sensors at key locations and by receiving signals from pedestrian wearable devices. This spatial distribution information mainly includes: crowd distribution location, density, and crowd movement speed.

[0038] Step 204: Construct a point queue prediction model based on spatial location distribution information and the objects to be evacuated.

[0039] In this embodiment, the Point-Queue Network Model (PQ) simplifies pedestrian movement in an evacuation crowd as a flow on a network while preserving key features such as queuing and delay. Specifically, areas and locations within the target building are mapped as nodes, and paths connecting these nodes are mapped as edges. A directed graph network is constructed based on spatial location distribution information as the topological structure, serving as the PQ network model. In the PQ network model, pedestrians on each edge l either move at free flow speed or queue at the tail of the edge at zero speed. The evacuation scenario is abstracted as a directed graph network. Where N is a set of nodes, mapping evacuation nodes such as rooms and staircases. Let edge be a set of edges, mapping the path or passage between two nodes. For any node... If there exists from arrive If the connection is direct, then a directed edge is defined. .definition: , indicating entering the node The set of all edges; , indicating leaving the node The set of all edges.

[0040] Step 205: Based on flood disaster data, spatial location distribution information, point queue prediction model, and the preset MPC model prediction and control framework, construct a prediction and control optimization model for the target building in the flood disaster scenario.

[0041] In this embodiment, the quantity change information and queue change information of the objects to be evacuated are extracted based on the spatial location distribution information; the MPC state space model is constructed based on the constraints corresponding to the quantity change information and queue change information; the corresponding nonlinear programming problem is solved based on the preset constraints corresponding to the nodes and edges in the point queue prediction model and the MPC state space model to obtain the predictive control optimization model.

[0042] The pre-defined MPC model predictive control framework employs an open-loop predictive control method with a rolling optimization strategy. This framework achieves optimal control of the controlled object by solving a finite-time optimization problem online. It comprises three main components: state acquisition, the MPC controller, and the controlled object, forming a complete control loop. The MPC controller, as the core component, includes three core parts: the predictive model, the optimal problem solver, and the objective function and constraint settings.

[0043] The predictive model describes the dynamic behavior of the controlled object, establishes the mathematical relationship between the current state of the system, the control input, and the future state, and provides state prediction capability for the optimization algorithm. The optimal problem solver is responsible for solving the predictive control optimization problem in each control cycle, and calculates the optimal control sequence in the future prediction time domain through the optimization algorithm. The objective function and constraint setting module defines the optimization objective and feasible region boundary based on the control performance requirements and system physical constraints.

[0044] In specific applications within flood disaster scenarios, the MPC framework employs an open-loop control sequence generation approach. This involves using a point queue prediction model to predict the future development of the flood situation based on flood data and spatial distribution information at the current moment, and generating an open-loop control sequence for the target building accordingly. The first control variable in this open-loop control sequence is output as the actual execution command to the controlled object at the current moment. At the start of the next control cycle, the system re-collects current state information and recalculates the optimal control sequence based on the new state data, thus achieving the rolling optimization characteristic of predictive control. Through this open-loop control method, the MPC framework can achieve predictive optimal control of the target building in flood disaster scenarios while satisfying system constraints.

[0045] Step 206: Obtain the evacuation prediction and control optimization results of the target building under the flood scenario through the predictive control optimization model and the hybrid solution framework based on the interior point method optimized by the velocity pause particle swarm algorithm.

[0046] In this embodiment, the velocity-pause particle swarm optimization algorithm is used to perform a global search in the entire feasible region of the MPC predictive control optimization model to obtain an initial solution. Using the precise gradient information of the objective function and constraint function, high-precision local optimization is performed starting from the initial solution provided by the particle swarm optimization algorithm, and the solution is quickly converged to the optimal solution that satisfies the KKT conditions, which is taken as the result of the evacuation predictive control optimization.

[0047] Step 207: Send evacuation instruction information to the objects to be evacuated based on the evacuation prediction and control optimization results.

[0048] Through the above process, evacuation prediction and control optimization results for the target building in a flood scenario are generated, and evacuation instruction information is generated based on the results and sent to the objects to be evacuated, so that the objects to be evacuated can evacuate according to the evacuation instruction information.

[0049] The evacuation path optimization method for flood scenarios provided in this invention establishes a predictive control optimization model based on flood dynamic parameters and personnel evacuation dynamics. The control results are obtained based on a gradient-information-based hybrid solution framework optimized using velocity-pause particle swarm optimization, ensuring the stability and reliability of the crowd evacuation path control results and improving the safety and effectiveness of personnel evacuation from buildings in flood scenarios.

[0050] In some optional embodiments of the present invention, such as Figure 4 As shown, the nodes in the PQ network model constructed in step 202 above can be divided into the following categories:

[0051] Virtual source node Virtual nodes have no physical meaning and are used to assist in constructing edges;

[0052] Entity Node These are nodes that actually correspond to architectural spaces such as rooms, and have practical significance. They are divided into starting point physical nodes that represent the true evacuation start point, exit physical nodes that represent the true evacuation end point, and ordinary physical nodes. Ordinary physical nodes can be further subdivided according to the required building type.

[0053] Virtual sink node Virtual nodes have no physical meaning and are used to assist in constructing edges.

[0054] The edges in this PQ network model can be categorized as follows:

[0055] Virtual Source Node - Starting Point Node: The edge connecting the virtual source node and the starting point entity. It does not represent the actual physical path, but is used to describe the initial distribution of people within the building at the start of the evacuation.

[0056] Entity Node-Entity Node: An edge connecting two different entity nodes. It represents the actual evacuation path from one location to another and has an actual length attribute;

[0057] Exit Node - Virtual Destination Node: Connects the physical exit node and the virtual destination node. It does not represent the actual physical path, but rather the process by which pedestrians reach the safe area via the evacuation endpoint.

[0058] Virtual sink node - exit node: Connects the virtual sink node and the physical exit node. It does not represent the actual physical path, but is used to indicate that when the queue time at a certain exit is too long, pedestrians are allowed to leave the path and change their destination to other exits.

[0059] In an evacuation network, if a passageway allows bidirectional pedestrian flow but is located in a narrow area, the convergence of opposing groups can easily create intersections, generating new congestion points and impacting evacuation efficiency. Therefore, constraints at narrow paths need to be specially considered during network construction. Represent narrow paths in the network, and mark all paths containing narrow passages according to the evacuation direction and add them to the set. This allows for the implementation of special flow constraints on narrow road sections during subsequent optimization control, preventing congestion caused by the intersection of two-way pedestrian traffic.

[0060] Therefore, in this PQ network model, the pedestrian's movement speed and state are also important parameters for evacuation prediction. In this embodiment, the flood scenario can be, for example, a flood situation, and the pedestrian's movement speed and state are determined based on empirical formulas obtained from existing underwater walking experiments and evacuation studies.

[0061] The pedestrian movement speed in a flood scenario is determined by the following formula:

[0062]

[0063] ,

[0064] in, Indicates water depth; Indicates the speed of water flow; This is the average speed of an adult walking freely on a dry surface, where a and b are constants: the free speed under conditions of use. In the case of flood-affected speed formula, the formula is used. , At this point, the critical specific flood force is approximately The corresponding water depth is approximately 4.9 cm.

[0065] The risk levels of flood impact are classified into the following four levels according to the UK Environment Agency (EA) classification criteria:

[0066] ,

[0067] Where v is the pedestrian's moving speed and h is the water depth.

[0068] Low risk: When At this time, pedestrians are in a low-flood-risk state. Under this state, the flood poses relatively little obstacle to pedestrian movement, pedestrians have a high degree of mobility, and can evacuate relatively smoothly.

[0069] Medium risk: If If the flood level is high, it is classified as a moderate flood risk. At this point, the impact of the flood on pedestrians begins to appear, and the speed of pedestrian evacuation may be somewhat slowed down, making movement more difficult.

[0070] High risk: When This indicates that pedestrians face a high risk of flooding. In a high-risk state, flooding poses a significant threat to pedestrians, severely restricting their movement and making evacuation more difficult and dangerous.

[0071] Highest risk: If Pedestrians are at the highest risk of flooding. In such extreme circumstances, floods can pose a direct threat to the lives of pedestrians, their movements are almost completely restricted, and evacuation is extremely difficult.

[0072] During flood evacuation, pedestrians face different risk levels and adjust their evacuation status according to the flood's danger level: if pedestrians are in low, medium, or high flood danger levels, they evacuate by slowing down to different degrees. Pedestrians in this state can improve their evacuation success rate through evacuation guidance and are included in the PQ network; among them, pedestrians in the high-risk state will instinctively abandon the current high-risk path and turn to a nearby path with a lower risk level; if pedestrians reach the highest risk level, they cannot leave the flood area on their own and must remain in place to wait for rescue. Pedestrians in this state are not included in the PQ network, but instead directly send a rescue signal to wait for the next rescue step.

[0073] In some optional embodiments of the present invention, in step 205, the quantity change information and queue change information of the objects to be evacuated are extracted based on the spatial location distribution information; the MPC state space model is constructed based on the constraints corresponding to the quantity change information and queue change information; the corresponding nonlinear programming problem is solved based on the preset constraints corresponding to the nodes and edges in the point queue prediction model and the MPC state space model to obtain the predictive control optimization model.

[0074] The objective function of this predictive control optimization model consists of the following parts:

[0075] a) Minimize the total number of people remaining at the end of the predicted state: This represents the total number of people remaining in the network at the end of the prediction time domain. This is the primary optimization objective, aiming to minimize the number of people remaining in the system at the end of the prediction time domain.

[0076] b) Distance penalty for remaining pedestrian terminal states: This penalty is calculated by multiplying the remaining number of people by the travel time of the road segment, ensuring that the remaining crowd is located as close to the exit as possible. This avoids suboptimal solutions that only consider minimizing the number of remaining people, such as situations where the remaining number of people decreases but they gather far from the exit, resulting in poor overall evacuation effectiveness.

[0077] c) Control costs by penalizing traffic transfers between nodes to prevent excessively frequent lane-switching and improve the stability of the control strategy.

[0078] d) To mitigate flood risks, consider the flood risk factors at each exit and prevent pedestrians from evacuating to high-risk areas. By penalizing the product of the number of people on each exit path and the risk coefficient of that exit, pedestrians are guided to prioritize evacuation routes with lower risks.

[0079] The constraints include:

[0080] Constraint 1. State dynamic equations

[0081] Used to describe the dynamic changes in the number of people on each road segment. It satisfies the basic flow conservation principle: the number of people on each road segment at the next moment is equal to the current number of people in that time period plus the inflow minus the outflow. : Indicates the time. Time Side The number of people on the platform. : Indicates the time. When entering the border The flow of people (inflow). : Indicates the time. From the side The outflow of people (outflow volume) is represented by u, which is the control variable in the MPC problem, x and λ are state variables, and v is an intermediate variable.

[0082] Constraint 2. Dynamic equation for queue length

[0083] This describes the changes in queue length within a road segment, used to simulate the movement of pedestrians along that segment. : Indicates the time. Time Side The number of people queuing at a point (queue length). The PQ network model assumes that pedestrians enter a road segment and pass through it at free-flow velocity, then leave the segment or join the queue at the end of the segment. Therefore, the queue length at each time step depends on the number of people queuing at the previous time step and the number of people queuing at that time step. Entering the link at any time Number of people at time as well as The outflow volume of this road segment at any given time. Among them, The outflow volume of a road segment at any given time is affected by the traffic capacity of that road segment. The impact.

[0084] The state-space model of MPC is constructed based on constraints 1 and 2 above. Through the above process, the quantity change information and queue change information of the objects to be evacuated are extracted based on the spatial location distribution information; the MPC state-space model is constructed based on the constraints corresponding to the quantity change information and queue change information.

[0085] Constraint 3. Node flow conservation

[0086] All nodes in the network maintain constant node traffic at all times. For non-sink nodes, the total inflow to the node equals the total outflow to the node. For sink (exit) nodes, inflow is allowed to exceed outflow, indicating that the population has left the network and reached a safe area.

[0087] Constraint 4. Narrow Path Constraint

[0088] Limit the total number of people on narrow, two-way routes in evacuation scenarios to prevent congestion caused by people traveling in opposite directions meeting in narrow passages. For each narrow route... and each time step The sum of pedestrian traffic in all directions on a narrow road segment should not exceed the capacity of that narrow road segment. The capacity of a narrow road segment is determined using a density-based capacity formula.

[0089] Constraint 5. Non-negativity constraint

[0090] Ensure that the state variables, queue length, inflow and outflow are all non-negative values.

[0091] In summary, the mathematical expression for the nonlinear programming problem in the predictive control optimization model is:

[0092] ,

[0093] ,

[0094] in, Let represent the set of all edges in the network, and let represent the passenger flow channels from the carriage to the exit and between the exits. The set of nodes in the network is denoted as and the set of nodes at the sink is denoted as . ; This represents the length of the model's prediction time domain and the number of time steps the MPC controller predicts forward. :time Time Side The number of pedestrians on the road; :time Time Side Queue length; Indicates time When entering the border Traffic; :time From the side Outflow of traffic; :side Maximum throughput capacity; Pedestrians crossing the edge Free walking time; This refers to a set of narrow road sections, including all road sections where capacity constraints need to be specially considered; ,and These represent entering the node respectively. The set of edges and nodes The set of edges from which the starting point is located; Indicates a narrow road section The capacity indicates the maximum allowable traffic flow on that road segment: Where v represents the average speed of pedestrians, ρ represents the critical density of pedestrians, and W represents the effective width of the road segment.

[0095] This is the terminal state distance penalty weight coefficient, used to balance the relationship between the two objectives of minimizing the number of terminals and optimizing the location of the remaining population; This represents the control cost weighting parameter, used to adjust the intensity of export transfer control; I is a risk weighting coefficient that adjusts the degree of influence of flood risk on evacuation decisions. i It is the set of all outgoing edges with exit i as the head node. Indicates export Flood risk value at the location:

[0096] ,

[0097] in, This indicates that the area at the outlet has an average flood strength higher than that of the surrounding area. This is the critical value for the safe evacuation of pedestrians during a flood.

[0098] exit At any moment The effective throughput capacity is determined by the following formula:

[0099] ,

[0100] Among them, M l (k) is the specific flood force at outlet l at the k-th time step; For export Basic traffic capacity: N is the maximum number of people passing through the exit per unit width per time step; is the width of the gate, and f is the capacity decay function.

[0101] The capacity attenuation function based on the specific flood force is defined as follows:

[0102] ,

[0103] in, Represents the sensitivity coefficient; This represents the maximum specific flood force threshold for pedestrian safe evacuation. The critical specific flood force value, representing the point at which flood action begins to significantly affect traffic capacity, and these two parameters together determine the attenuation function. Boundary conditions and variation characteristics.

[0104] Through the above process, based on the preset constraints of the nodes and edges in the point queue prediction model and the MPC state space model, the corresponding nonlinear programming problem is solved to obtain the predictive control optimization model.

[0105] Please refer to Figure 8 , Figure 8 A flowchart illustrating an evacuation route optimization method in a flood scenario provided by this disclosure embodiment, specifically for... Figure 2 Step 206 in process 200 provided a specific implementation. Other steps in process 200 are not adjusted; a new complete embodiment is obtained by replacing step 206 with the specific implementation provided in this embodiment. Process 700 includes the following steps:

[0106] Step 701: Perform a global search across the entire feasible region of the predictive control optimization model using the velocity paused particle swarm optimization algorithm (VPPSO algorithm) to obtain an initial solution.

[0107] In this embodiment, the above process mainly includes:

[0108] Step 1: Randomly generate the velocity and position vectors of each particle in the particle swarm within a preset range. At the start of the particle swarm optimization process, the velocity and position of each particle are randomly generated within a specific range:

[0109] ,

[0110] in, and Particles velocity vector and position vector Let be the dimension. For group size.

[0111] Step 2: Based on the velocity vector and position vector, update the first velocity and first position of the particles in the first group of the particle swarm using the particle swarm optimization algorithm improved by pausing based on velocity.

[0112] During particle swarm iteration, particles The best global particle discovered to date and its best location Update its speed and position respectively:

[0113] ,

[0114] The particle iterative update formula is as follows:

[0115] ,

[0116] in, For inertial weights, and These are the cognitive acceleration coefficient and the social acceleration coefficient, respectively. and Let be two random variables that are uniformly distributed in the range [0,1].

[0117] Among them, inertia weight Its purpose is to avoid the velocity explosion problem faced by the standard particle swarm optimization algorithm; acceleration coefficient and Control the particle direction separately and The speed. These three PSO (Particle Swarm Optimization) parameters ( It plays a crucial role in balancing PSO exploration and development capabilities.

[0118] After updating the particle's velocity and position, its optimal individual position is updated as follows:

[0119] ,

[0120] Only when newly generated particles Its fitness is better than The particle will only be updated when its current fitness is reached. The best position for an individual.

[0121] The global optimal position has been updated as follows;

[0122] ,

[0123] Based on the velocity pause strategy, particles are allowed to move at the same constant velocity as in the previous iteration, calculated using the following formula:

[0124] ,

[0125] in, It's a speed pause parameter. A value less than 1 indicates that the speed pause strategy is enabled, while a value greater than 1 indicates that the classic particle swarm optimization algorithm is used. These are random numbers that are uniformly distributed in the range [0,1].

[0126] To further help PSO avoid premature convergence, the velocity equation of the traditional PSO algorithm is modified by changing the first velocity term and omitting its inertial weight component, as follows:

[0127] ,

[0128] in, Let be three random variables uniformly distributed in the range [0,1]. The mathematical expression is as follows:

[0129] ,

[0130] Where b is a constant, T is the maximum number of iterations, and t is the current number of iterations.

[0131] By applying the concept of velocity pause and utilizing the modified velocity equation in the above formula, the particles in VPPSO update their velocities as follows:

[0132] ,

[0133] particle The location has been updated as follows:

[0134] ,

[0135] In this embodiment, to maintain population diversity and avoid premature convergence, the proposed algorithm of the VPPSO algorithm will use the total population... They are divided into two groups. The first group consists of... The particles in these examples are composed of classical particle swarm optimization mechanisms that use the aforementioned velocity-pause strategy to update their velocity and position.

[0136] Step 3: Update the second position of the particles in the second group of the particle swarm based on the historical global best position.

[0137] In this embodiment, the second group only uses To update their positions. The update rule for each particle in the second population is as follows:

[0138] ,

[0139] in, Let be two random variables that are uniformly distributed in the range [0,1].

[0140] Step 4: Based on the first velocity, first position, second position, and preset inertia weights, cognitive acceleration coefficients, and social acceleration coefficients, obtain the fitness distribution of the particle swarm. Building upon the above process, evaluate the fitness performance of all particles, dynamically update the individual optimal position and the global optimal position, and prevent the algorithm from prematurely converging to a local optimum.

[0141] Step 5: Update the global optimal position of the particle swarm based on the fitness distribution until the preset termination condition is met, and obtain the potential optimal solution set.

[0142] In this embodiment, the main advantage of the velocity pause strategy is the addition of a third movement option (constant velocity), which helps balance exploration and development and avoids the severe premature convergence of classic PSO. The VPPSO optimization process begins by randomly generating the velocities and positions of all particles. During the VPPSO iteration, particles in the first swarm update their velocities and positions according to the velocity pause improved particle swarm optimization algorithm. Particles in the second swarm update their positions only based on the simplified global optimum update. Then, the fitness of all particles is evaluated, and the individual best positions of the entire swarm are updated. If a better fitness is obtained in any swarm, the VPPSO process is repeated based on the updated global best position until the stopping condition is met, a potential high-quality solution region is found, and a potential optimal solution set is obtained.

[0143] Step 702: Using the precise gradient information of the preset objective function and preset constraint function, perform local optimization on the initial solution to obtain the optimal solution that satisfies the Karl von Kühn-Tucker conditions (KKT conditions).

[0144] In this embodiment, the above process mainly includes:

[0145] Step 1: Combine slack variables to transform the constraints into equality constraints;

[0146] Introduced slack variable s: Transform inequality constraints into equality constraints:

[0147] ,

[0148] in This refers to all the inequality constraints, such as flow capacity constraints, non-negativity constraints, and network capacity constraints, mentioned in all the above nonlinear programming problems.

[0149] Step 2: Combine the logarithmic barrier function to handle the non-negativity constraint of the slack variables;

[0150] By introducing a logarithmic barrier function to handle the non-negativity constraint of the slack variables, the original problem is transformed into a barrier subproblem:

[0151] ,

[0152] ,

[0153] in, This represents constraints such as crowd flow conservation and queue length updates. For obstacle parameters, s i It is the i-th component of the introduced slack variable, and f(u) represents the objective function of the original linear programming problem.

[0154] Step 3: Construct the Lagrangian function based on the equality constraints and non-negativity constraints, and establish the Carlow-Kuhn-Tucker first-order necessary condition system;

[0155] The Lagrangian function is constructed and the necessary conditions of the first order KKT are established as follows:

[0156]

[0157]

[0158]

[0159]

[0160]

[0161] in, It is a diagonal matrix of slack variables. It is a diagonal matrix with dual variables. It is a unit vector, η represents the equality constraint Lagrange multiplier, and z is the inequality constraint Lagrange multiplier. Complementarity condition. Ensure that the interior point method iterates within the feasible region.

[0162] Step 4: Perform Newtonian linearization on the Caro-Kun-Tucker first-order necessary condition system to obtain a symmetric linear system;

[0163] Solving for the KKT first-order necessary condition system using Newton's direction and the Mehrotra strategy first involves Newton linearizing the KKT conditions to obtain a symmetric linear system:

[0165] ,

[0166] in: Let be the Hessian matrix of the Lagrange function with respect to the control variable u. Let be the Jacobian matrix of the equality constraints with respect to u. Let be the Jacobian matrix of the inequality constraints with respect to u. This indicates the Newton search direction for each variable.

[0167] Step 5: Obtain the predicted direction of the affine system based on the symmetric linear system;

[0168] The Mehrotra prediction-correction strategy is employed for high convergence performance. The steps are as follows: Solve... The affine system obtains the predicted direction In this context, the superscript aff represents the affine prediction step;

[0169] Step 6: Calculate the affine step size of the original variable and the affine step size of the dual variable;

[0170] Specifically, the affine step size of the original variable and the affine step size of the dual variable are calculated using the following formula:

[0171] ,

[0172] Where pri represents the original variable and dual represents the dual variable;

[0173] Step 7: Update the centered parameter and the barrier parameter of the logarithmic barrier function based on the affine step size of the original variable and the affine step size of the dual variable;

[0174] Specifically, the centralization parameter and the new barrier parameter are updated using the following formula:

[0175] ,

[0176] in, is the centering coefficient, and p is the number of inequality constraints.

[0177] Step 8: Correct the symmetric linear system based on the updated centering parameters and barrier parameters; Step 9: Obtain the final search direction based on the corrected symmetric linear system; Step 10: Obtain the optimal solution that satisfies the Caro-Kuhn-Tucker conditions based on the final search direction.

[0178] Based on the above process, the corrected Newtonian system is solved to obtain the final search direction.

[0179] Line search updates and convergence determination employ backtracking line search to ensure the Armijo condition, and calculate the feasible step size:

[0180] ,

[0181] Update iteration points:

[0182] ,

[0183] ,

[0184] Among them, 0.995 is a safety factor that takes a commonly used value to ensure strict feasibility; It is the step size of the original variable. The step size is the dual variable, and k is the number of iterations.

[0185] Its convergence determination is based on the following three criteria:

[0186] 1. Duality feasibility: ,

[0187] 2. Initial feasibility: ,

[0188] 3. Complementary gap: ,

[0189] in These represent the duality feasibility tolerance, the original feasibility tolerance, and the complementarity tolerance, respectively, while p represents the number of inequality constraints. It is the minimum obstacle parameter value. Represents the infinite norm, It is the inner product of the slack variable and the dual variable.

[0190] During the calculation process, for the starting point of the calculation, candidate solutions provided by VPPSO are used as initial values ​​for a warm start, and adaptive barrier parameters are used for updating:

[0191] ,

[0192] Here, θ is the obstacle parameter reduction factor, θ∈(0,1). Through the above-described local accurate solution process using the interior point method, it can quickly converge to a high-precision optimal control sequence within the optimal region determined by VPPSO. This hybrid solution framework significantly improves the convergence efficiency of the crowd evacuation network optimization problem and can achieve fast and accurate solution of the real-time optimal control sequence in the scenario of flooding in building space.

[0193] In some optional embodiments of the present invention, the predictive control optimization implementation process refers to three main parts: the data acquisition and estimation process of the evacuation control method based on the MPC control prediction model, the open-loop optimal control process, and the control implementation feedback process, including:

[0194] 1. Data Acquisition and Estimation Process

[0195] The spatial distribution information of people to be evacuated within the current building space is collected by setting up pedestrian number sensors at key locations and receiving signals from pedestrian wearable devices. A mathematical expression for the current evacuation situation is obtained by combining the collected pedestrian distribution information with a PQ network model. Based on flood data, building space layout information, and the PQ network model, a predictive control optimization model for the building space under flood scenarios is established and solved.

[0196] 2. Open-loop optimal control process

[0197] The MPC controller solves the prediction time domain based on the current environment and network states. The optimization problem aims to determine the optimal crowd distribution strategy among the edges. The controller's inputs include the number of people on each edge, queue length, and flood risk and capacity at each exit. In the open-loop optimal control calculation, the controller first predicts the future evacuation state evolution based on a PQ network model, and then constructs an optimization problem with the objective of minimizing the number of unevacuated people and the associated risks at all discrete time points within the predicted time domain. By solving this optimization problem, the open-loop control sequence within the predicted time domain is obtained. ,in Indicates at time When entering the border The flow of people (inflow).

[0198] 3. Controlling the feedback process

[0199] To avoid the inevitable discrepancy between the predictive model and the actual system dynamics, which would prevent the actual system evolution from perfectly following the predicted optimal trajectory, it is necessary to remeasure the new system state after a specific control cycle and use it as a new initial condition for the predictive model. This allows for the regeneration of an open-loop control law suitable for the next stage of evacuation, and the law is executed only for the current moment in each control cycle. Control actions This process is repeated until the evacuation is complete, constituting a rolling optimization process with feedback.

[0200] The alienation control strategy for implementing predictive control optimization strategies, specifically the operational methods of the entire optimization method in the actual scenario, includes:

[0201] Control signal conversion: The output of the predictive control optimization model is converted into standardized pedestrian control signals that are easy to implement for pedestrians in the building space. The pedestrian control signals include evacuation direction indication signals, movement speed control signals, path selection guidance signals, and assembly point indication signals.

[0202] Signal application steps: The pedestrian control signal is applied to the crowd to be evacuated in real time through audio-visual display equipment, voice broadcasting system or mobile terminal to ensure the effective transmission of control instructions;

[0203] Cyclic control steps: The data acquisition, model optimization and control implementation process is repeated at preset time intervals. The preset time interval is dynamically adjusted according to the building size and the urgency of evacuation, and the value range is 10 seconds to 120 seconds.

[0204] Status monitoring steps: Real-time monitoring of pedestrian evacuation progress and the distribution of people in the space, identifying those who have successfully transitioned during the evacuation process and those who are unable to evacuate on their own.

[0205] Termination decision step: The control process terminates when one of the following conditions is met:

[0206] a) All pedestrians were successfully evacuated to a safe area;

[0207] b) Detection of pedestrians arriving in a state of waiting for rescue who are unable to evacuate on their own;

[0208] c) The system detected an irresistible environmental obstacle;

[0209] Signal transmission steps: Based on the evacuation results, transmit evacuation success confirmation signals or emergency rescue request signals to relevant systems or personnel, and record evacuation process data for subsequent analysis.

[0210] In the embodiments of the present invention, the evacuation route optimization method for flood scenarios described in any of the above embodiments can be applied to, for example... Figure 5 The simulation scenario depicts pedestrian evacuation within the target building. Details of the number of people moving between exits within 360 seconds on the platform level (3rd basement level) in this scenario are as follows: Figure 5 As shown. Corresponding to Figure 5 The simulation scene shown Figure 6 The figure shows the change in the number of evacuees over time achieved by applying the evacuation route optimization method of this invention. Figure 7The figure shows the change in the number of evacuees over time without applying the evacuation route optimization method of this invention. The comparison shows that applying the evacuation route optimization method of this invention can achieve more efficient evacuation.

[0211] Further reference Figure 9 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of an evacuation route planning device for flood scenarios. This device embodiment is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0212] like Figure 9 As shown, the evacuation route planning device 800 in the flood scenario of this embodiment may include: a building parameter acquisition module 801, a flood parameter acquisition module 802, a spatial distribution information acquisition module 803, a point queue prediction model construction module 804, a prediction control optimization model construction module 805, an evacuation prediction control optimization result module 806, and an evacuation instruction sending module 807. The system includes the following modules: a building parameter acquisition module 801, configured to acquire the building space parameters of the target building; a flood parameter acquisition module 802, configured to perform hydrodynamic simulations of a preset flood scenario based on the building space parameters to obtain flood data; a spatial distribution information acquisition module 803, configured to acquire the spatial location distribution information of objects to be evacuated within the space of the target building; a point queue prediction model construction module 804, configured to construct a point queue prediction model based on the spatial location distribution information and the objects to be evacuated; a predictive control optimization model construction module 805, configured to construct a predictive control optimization model for the target building under the flood scenario based on flood data, spatial location distribution information, the point queue prediction model, and a preset MPC model predictive control framework; an evacuation predictive control optimization result module 806, configured to obtain the evacuation predictive control optimization result of the target building under the flood scenario through the predictive control optimization model and a hybrid solution framework based on the interior point method optimized using the velocity-pause particle swarm algorithm; and an evacuation instruction sending module 807, configured to send evacuation instruction information to the objects to be evacuated based on the evacuation predictive control optimization result.

[0213] In this embodiment, the specific processing and technical effects of the building parameter acquisition module 801, flood parameter acquisition module 802, spatial distribution information acquisition module 803, point queue prediction model construction module 804, prediction control optimization model construction module 805, evacuation prediction control optimization result module 806, and evacuation instruction sending module 807 in the evacuation route planning device 800 under flood scenarios can be referred to respectively. Figure 2 The relevant descriptions of steps 201-207 in the corresponding embodiments will not be repeated here.

[0214] This embodiment exists as a device embodiment corresponding to the above method embodiment. The evacuation path planning device in the flood scenario provided in this embodiment establishes a predictive control optimization model based on flood dynamic parameters and personnel evacuation dynamics. The control result is obtained based on a gradient information-based hybrid solution framework optimized by velocity-pausing particle swarm optimization, which ensures the stability and reliability of the crowd evacuation path control result and improves the safety and effectiveness of personnel evacuation in buildings in the flood scenario.

[0215] According to embodiments of this disclosure, this disclosure also provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to implement the evacuation route optimization method in the flood scenario described in any of the above embodiments.

[0216] According to embodiments of this disclosure, this disclosure also provides a readable storage medium storing computer instructions that enable a computer to implement the evacuation route optimization method for flood scenarios described in any of the above embodiments when executed.

[0217] According to embodiments of this disclosure, this disclosure also provides a computer program product that, when executed by a processor, can implement the evacuation route optimization method for flood scenarios described in any of the above embodiments.

[0218] Figure 10 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0219] like Figure 10As shown, device 900 includes a computing unit 901, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 902 or a computer program loaded from storage unit 908 into random access memory (RAM) 903. RAM 903 may also store various programs and data required for the operation of device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / output (I / O) interface 905 is also connected to bus 904.

[0220] Multiple components in device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, etc.; output unit 907, such as various types of monitors, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0221] The computing unit 901 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as the evacuation route optimization method in a flood scenario. For example, in some embodiments, the evacuation route optimization method in a flood scenario can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the evacuation route optimization method in a flood scenario described above can be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform an evacuation route optimization method in a flood scenario by any other suitable means (e.g., by means of firmware).

[0222] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0223] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0224] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0225] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0226] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0227] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0228] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0229] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for optimizing evacuation path in a flood scenario, characterized in that, The method comprises: acquiring building space parameters of a target building; performing water dynamics simulation on a preset flood scenario based on the building space parameters to obtain flood data; collecting spatial position distribution information of objects to be evacuated in the target building; constructing a point queue prediction model based on the spatial position distribution information and the objects to be evacuated; constructing a predictive control optimization model for the target building under the flood scenario based on the flood data, the spatial position distribution information, the point queue prediction model, and a preset MPC model predictive control framework; obtaining an evacuation predictive control optimization result of the target building under the flood scenario through the predictive control optimization model and a hybrid solution framework based on an interior point method optimized by a speed-halt particle swarm algorithm; sending evacuation instruction information to the objects to be evacuated based on the evacuation predictive control optimization result.

2. The method of claim 1, wherein, The building space parameters are structure data of the target building, The water dynamics simulation on the preset flood scenario based on the building space parameters to obtain the flood data comprises: establishing a three-dimensional simulation model of the target building under the flood scenario based on the structure data and a preset water dynamics model; performing flood scenario simulation on the target building based on the three-dimensional simulation model, and collecting the flood data in the whole flood stage, wherein the flood data comprises water depth, flow velocity, and inflow rate.

3. The method of claim 1, wherein, The construction of the point queue prediction model based on the spatial position distribution information and the objects to be evacuated comprises: mapping the regions and position points in the target building into nodes, mapping the paths connecting the nodes into edges, taking the spatial position distribution information as the basis of the topological structure, and constructing a directed graph network based on the point queue prediction model as the point queue prediction model.

4. The method of claim 3, wherein, The construction of the point queue prediction model based on the spatial position distribution information and the objects to be evacuated further comprises: determining the risk level of the objects to be evacuated based on the corresponding relationship between the moving speed of the objects to be evacuated and the water depth and flow velocity; counting the objects to be evacuated into the point queue prediction model based on the risk level.

5. The method of claim 1, wherein, The construction of the predictive control optimization model for the target building under the flood scenario based on the flood data, the spatial position distribution information, the point queue prediction model, and the preset MPC model predictive control framework comprises: extracting quantity change information and queue change information of the objects to be evacuated based on the spatial position distribution information; constructing an MPC state space model based on the constraint conditions corresponding to the quantity change information and the queue change information; solving a corresponding nonlinear programming problem based on the preset constraint conditions corresponding to the nodes and edges in the point queue prediction model and the MPC state space model to obtain the predictive control optimization model.

6. The method of claim 5, wherein, The preset MPC model predictive control framework comprises: The open-loop predictive control method is constructed based on a rolling optimization strategy, and the open-loop predictive control method comprises a control cycle formed by state acquisition, an MPC controller and a controlled object; the MPC controller comprises a prediction model, an optimal problem solver and a target function and constraint condition setting module; The prediction model in the MPC controller is used to establish a mathematical relationship between the current state of the system, the control input and the future state, and to provide state prediction capability for the optimization algorithm; The optimal problem solver is used to solve the predictive control optimization problem in each control period, and the optimal control sequence in the future prediction time domain is calculated through the optimization algorithm; The target function and constraint condition setting module is used to define the optimization target and the feasible region boundary according to the control performance requirements and the system physical limitation conditions; Based on the flood data and the spatial position distribution information, an open-loop control action sequence for the target building is generated by using a point queue prediction model, and state information is re-acquired in each control period to realize rolling optimization control.

7. The method of claim 5, wherein, The evacuation predictive control optimization result of the target building in the flood scenario is obtained by using the mixed solving framework based on the interior point method and the velocity suspension particle swarm optimization algorithm optimized by the prediction control optimization model, and the evacuation predictive control optimization result comprises: The velocity suspension particle swarm optimization algorithm is used to perform global search in the entire feasible region of the prediction control optimization model to obtain an initial solution; The initial solution is locally optimized by using the exact gradient information of the preset target function and the preset constraint function to obtain an optimal solution satisfying the Karush-Kuhn-Tucker condition.

8. The method of claim 7, wherein, The velocity suspension particle swarm optimization algorithm is used to perform global search in the entire feasible region of the prediction control optimization model to obtain an initial solution, and the method comprises: Velocity vectors and position vectors of each particle in the particle swarm are randomly generated in a preset range; Based on the velocity vectors and the position vectors, the first velocity and the first position of the particles in a first population in the particle swarm are updated by using the particle swarm optimization algorithm improved according to the velocity suspension; The second position of the particles in a second population in the particle swarm is updated based on the historical global optimal position; Based on the first velocity, the first position, the second position and preset inertia weights, cognitive acceleration coefficients and social acceleration coefficients, a fitness distribution of the particle swarm is obtained; The global optimal position of the particle swarm is updated based on the fitness distribution until a preset termination condition is satisfied, and a potential optimal solution set is obtained.

9. The method of claim 7, wherein, The initial solution is locally optimized by using the exact gradient information of the target function and the constraint function through the gradient solver to obtain an optimal solution satisfying the Karush-Kuhn-Tucker condition, and the method comprises: The constraint conditions are converted into equality constraint conditions in combination with slack variables; The non-negativity constraint of the slack variable is processed in combination with a logarithmic barrier function; A Lagrange function is constructed based on the equality constraint conditions and the non-negativity constraint, and a Karush-Kuhn-Tucker first-order necessary condition system is established; The Karush-Kuhn-Tucker first-order necessary condition system is subjected to Newton linearization processing to obtain a symmetric linear system; The prediction direction of an affine system is obtained based on the symmetric linear system; computing a primal variable affine step and a dual variable affine step; updating a centering parameter and a barrier parameter of the log-barrier function based on the primal variable affine step and the dual variable affine step; correcting the symmetric linear system based on the updated centering parameter and barrier parameter; obtaining a final search direction based on the corrected symmetric linear system; obtaining an optimal solution satisfying the Karush-Kuhn-Tucker condition based on the final search direction.

10. A device for optimizing evacuation routes in a flooding scenario, characterized in that, comprise: a building parameter acquisition module configured to acquire building space parameters of a target building; a flood parameter acquisition module configured to perform hydrodynamic simulation on a preset flood scenario based on the building space parameters to obtain flood data; a spatial distribution information collection module configured to collect spatial position distribution information of objects to be evacuated within the target building; a grouping queue point queue prediction model construction module configured to construct a point queue prediction model based on the spatial position distribution information and the objects to be evacuated; a prediction control optimization model construction module configured to construct a prediction control optimization model for the target building under the flood scenario based on the flood data, the spatial position distribution information, the point queue prediction model, and a preset MPC model prediction control framework; an evacuation prediction control optimization result module configured to obtain an evacuation prediction control optimization result of the target building under the flood scenario through the prediction control optimization model and a hybrid solution framework based on an interior point method optimized using a velocity-halt particle swarm optimization algorithm; an evacuation instruction sending module configured to send evacuation instruction information to the objects to be evacuated based on the evacuation prediction control optimization result. 11.An electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the evacuation path optimization method under a flood scenario according to any one of claims 1-9.

Images