An activity facility automatic arrangement optimization method based on spatial topology constraint

Abstract

This invention provides an automatic layout optimization method for activity facilities based on spatial topological constraints, belonging to the field of automated layout technology. It includes: acquiring site and facility data to construct a spatial topological constraint model; and performing multi-objective iterative optimization on a set of candidate layout schemes after initial coordinate allocation. During optimization, the coordinate offset vectors of target activity facilities whose locations change are recorded in real time. Based on the preset rigid constraint distance and elastic potential function of the topological constraint edges, the cooperative displacement vectors of subordinate activity facilities are calculated, and linked updates are performed until convergence. This invention effectively curbs the destruction of spatial dependence by iterative mutations, significantly improving the algorithm's optimization efficiency and the engineering feasibility of the optimal layout scheme.

Description

Technical Field

[0001] This invention relates to the field of automated layout technology, and in particular to an automatic layout optimization method for activity facilities based on spatial topology constraints. Background Technology

[0002] In the preparation process for events such as commercial exhibitions, large-scale sporting events, and outdoor performances, the layout planning of temporary facilities within the event venue is a crucial step. These temporary facilities typically include a main stage, audience seating, security checkpoints, evacuation routes, and temporary power distribution boxes. Currently, the layout of such facilities relies heavily on the engineering experience of designers, utilizing 2D computer-aided design software for manual trial-and-error planning. In recent years, some automated layout methods have attempted to incorporate heuristic algorithms to assist in layout. For example, Chinese invention patent document CN110851962A discloses a "Digital Workshop Layout Optimization Method," which establishes a mathematical model that minimizes logistics costs and maximizes area utilization, and uses an improved adaptive genetic algorithm to solve the model, outputting a layout scheme for workshop equipment. This type of method based on traditional swarm intelligence algorithms has been initially applied in the layout of standard parts or in solid-state industrial scenarios.

[0003] As spatial planning demands become increasingly complex, research and development in the field of facility layout optimization is evolving from single-objective geometric interference verification to multi-objective comprehensive efficiency optimization. Current trends focus on introducing multi-objective swarm intelligent optimization algorithms or deep reinforcement learning models. By constructing a multi-dimensional fitness evaluation function within a defined three-dimensional feasible domain, candidate layout schemes are iteratively screened. This approach aims to obtain a comprehensive optimal solution across multiple dimensions, including pedestrian evacuation efficiency, visual experience, and pipeline deployment costs, thereby enhancing the automation of the layout process.

[0004] However, when facing specific scenarios with highly dynamic and interconnected features, such as event facilities, existing layout optimization algorithms suffer from a lack of underlying constraints. Traditional algorithms typically abstract each event facility into an independent geometric bounding box and perform random initialization and mutation iterations within a given grid area. This mechanism severs the inherent spatial topological constraints between event facilities. In practical engineering, the main stage and the sound control console need to satisfy an unobstructed line-of-sight cone constraint, security nodes need to satisfy an azimuth constraint of being adjacent to and perpendicular to the main pedestrian flow path, and the substation and high-energy-consuming facilities need to satisfy a physical distance constraint for safety cables. Because existing algorithms lack prior knowledge of these spatial topological dependencies, they are prone to disrupting the structure of facility groups with strong functional coupling when performing cross operations or pose mutation operations, thereby generating a large number of invalid solutions that violate fire safety regulations or functional logic. These defects result in low search efficiency and slow convergence speed in the vast solution space. The final output layout scheme lacks engineering feasibility and still requires manual secondary adjustments and repairs. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide an automatic arrangement optimization method for activity facilities based on spatial topology constraints. This invention solves the problem that traditional optimization algorithms in the prior art lack prior knowledge of the spatial topology constraints between facilities, resulting in low algorithm search efficiency and a lack of engineering feasibility in the output scheme.

[0006] To achieve the above objectives, the present invention provides the following solution: An automatic layout optimization method for activity facilities based on spatial topology constraints includes: Obtain environmental spatial data of the target site and a set of activity facilities to be arranged, and construct a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities; Within the three-dimensional feasible domain defined by the environmental spatial data, the set of activity facilities is initially assigned coordinates to obtain a set of candidate layout schemes; A multi-objective optimization algorithm is invoked to iteratively optimize the candidate layout scheme set. During the iterative optimization process, the target activity facilities that have changed positions in the candidate layout scheme set are extracted in real time, and the coordinate offset vectors of the target activity facilities are recorded. Based on the spatial topology constraint model, the target facility node corresponding to the target activity facility is located, and the subordinate activity facilities that have a physical dependency relationship with the target activity facility are extracted according to the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. Read the preset rigid constraint distance of the target topological constraint edge, and calculate the cooperative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector, and the preset elastic potential energy function; The spatial coordinates of the subordinate activity facilities are updated in a coordinated manner based on the coordinated displacement vector until the iterative optimization reaches the preset convergence condition, and the optimal arrangement scheme is obtained from the set of candidate arrangement schemes that have completed the coordinated update.

[0007] An automatic layout optimization system for activity facilities based on spatial topology constraints includes: The data acquisition and model building module is used to acquire environmental spatial data of the target site and a set of activity facilities to be arranged, and to build a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities. The initial coordinate allocation module is used to perform initial coordinate allocation on the set of activity facilities within the three-dimensional feasible domain defined by the environmental spatial data, so as to obtain a set of candidate layout schemes. The iterative optimization and offset recording module is used to call a multi-objective optimization algorithm to iteratively optimize the candidate layout scheme set, and during the iterative optimization process, extract the target activity facilities that have changed positions in the candidate layout scheme set in real time, and record the coordinate offset vector of the target activity facilities. The topology dependency extraction module is used to locate the target facility node corresponding to the target activity facility based on the spatial topology constraint model, and extract the subordinate activity facilities that have a physical dependency relationship with the target activity facility based on the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. The collaborative displacement calculation module is used to read the preset rigid constraint distance of the target topological constraint edge, and calculate the collaborative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector and the preset elastic potential energy function. The linkage update and scheme output module is used to perform linkage update on the spatial coordinates of the subordinate activity facilities based on the collaborative displacement vector until the iterative optimization reaches the preset convergence condition, and to obtain the optimal arrangement scheme from the candidate arrangement scheme set that has completed the linkage update.

[0008] The present invention discloses the following technical effects: This invention provides an automatic layout optimization method for activity facilities based on spatial topological constraints. By constructing a spatial topological constraint model representing the spatial dependencies between facilities, during the iterative process of a multi-objective optimization algorithm, once a target activity facility experiences a coordinate shift, the collaborative displacement vector of its subordinate activity facilities with which it has a physical dependency can be accurately calculated based on the rigid constraint distance of the topological constraint edge, the coordinate shift vector, and the elastic potential energy function, and a linked update can be performed. This linked update mechanism effectively curbs the disruption of the inherent functional coupling relationship of the facility group caused by traditional blind mutation operations, significantly reduces the probability of generating invalid solutions that violate physical rules and safety specifications, and significantly improves the search efficiency and convergence speed of the multi-objective optimization algorithm in the vast solution space. Simultaneously, it ensures that the final output optimal layout scheme is directly engineering-feasible, completely overcoming the shortcomings of existing technologies where layout schemes are prone to failure and highly dependent on manual secondary repair. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A flowchart of an automatic layout optimization method for activity facilities based on spatial topology constraints is provided for an embodiment of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0013] like Figure 1 As shown, this invention provides an automatic arrangement optimization method for activity facilities based on spatial topology constraints, including: Step 100: Obtain the environmental spatial data of the target site and the set of activity facilities to be arranged, and construct a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities; Step 200: Within the three-dimensional feasible domain defined by the environmental spatial data, perform initial coordinate assignment on the set of activity facilities to obtain a set of candidate layout schemes; Step 300: Call the multi-objective optimization algorithm to iteratively optimize the candidate layout scheme set, and during the iterative optimization process, extract the target activity facilities that have changed positions in the candidate layout scheme set in real time, and record the coordinate offset vector of the target activity facilities; Step 400: Based on the spatial topology constraint model, locate the target facility node corresponding to the target activity facility, and extract the subordinate activity facilities that have a physical dependency relationship with the target activity facility according to the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. Step 500: Read the preset rigid constraint distance of the target topological constraint edge, and calculate the cooperative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector and the preset elastic potential energy function; Step 600: Perform a linkage update on the spatial coordinates of the subordinate activity facility based on the collaborative displacement vector until the iterative optimization reaches the preset convergence condition, and obtain the optimal layout scheme from the candidate layout scheme set that has completed the linkage update.

[0014] Furthermore, the specific implementation process of step 100 is as follows: This embodiment performs the steps of acquiring environmental spatial data of the target site and the set of activity facilities to be arranged. This embodiment uses a LiDAR scanning device to acquire a three-dimensional point cloud array of the target site as the environmental spatial data, and performs voxel mesh filtering on the three-dimensional point cloud array to remove outliers and noise. The filtered scanning spatial resolution is accurate to 0.05 meters, used to characterize the three-dimensional terrain and unusable obstacle outlines of the site. Simultaneously, this embodiment reads a total of 120 independent facility units, including the main stage, power distribution box, and audience seating, from a pre-set activity preparation configuration library to form the set of activity facilities to be arranged; the activity preparation configuration library refers to a relational database established in advance by inputting the factory physical specifications of each facility. Subsequently, this embodiment extracts the association attributes of each activity facility in the set of activity facilities. These association attributes are metadata tags pre-set in the activity preparation configuration library, specifically used to characterize the mandatory engineering coupling requirements between different activity facilities in terms of physical connection distance, spatial orientation angle, and line-of-sight obstruction.

[0015] This embodiment abstracts each activity facility in the activity facility set into a corresponding facility node. This embodiment calls the spatial envelope algorithm to calculate the three-dimensional projection extrema of the physical contours of each activity facility's edge along the three orthogonal axes of the site environment coordinate system, thereby constructing an outer spatial bounding box containing the entire volume of the activity facility. Subsequently, this embodiment extracts the three-dimensional coordinate maxima and minima of the outer spatial bounding box. By summing the coordinate maxima and minima along the same coordinate axis and dividing by a constant two, the geometric centroid coordinates of the outer spatial bounding box are calculated. This embodiment uses these geometric centroid coordinates as the three-dimensional absolute coordinates of the corresponding abstracted facility node, storing them in double-precision floating-point format to retain the data precision of the three-dimensional coordinates to three decimal places. The facility node refers to a data structure that has completely stripped away the physical form of the activity facility, retaining only the three-dimensional absolute coordinates and the metadata tags, to serve as a discrete vertex in the subsequent topology network.

[0016] This embodiment establishes virtual connection edges between the facility nodes based on the associated attributes. If the associated attributes include a maximum cable connection threshold, this embodiment establishes virtual connection edges between the corresponding facility nodes that satisfy the physical distance measurement association rule. The physical distance measurement association rule is a judgment logic that forcibly limits the absolute value of the spatial straight-line distance between two facility nodes to not exceeding the safe power supply distance. This embodiment extracts the three-dimensional spatial absolute coordinates of the power distribution facility node and the high-energy-consuming facility node. By calculating the sum of the squares of the coordinate differences of the two sets of coordinate points on the three orthogonal spatial axes, and performing a square root operation on the sum of squares, the actual spatial Euclidean distance between the two nodes is obtained. This embodiment compares the actual spatial Euclidean distance with the maximum cable connection threshold extracted from the power safety specifications and set to 500 meters. When the actual spatial Euclidean distance is less than or equal to 500 meters, a virtual connection edge carrying a gravitational potential energy attribute is established between the power distribution facility node and the high-energy-consuming facility node. The gravitational potential energy attribute is used to provide a numerical weight for pulling inward when a position exceeds the boundary in subsequent multi-objective optimization.

[0017] When handling spatial orientation constraints, if the associated attribute includes the normal angle of the main road, this embodiment establishes a virtual connection edge that satisfies the orientation alignment association rule between the corresponding facility nodes. This embodiment generates a three-dimensional spatial orientation vector representing the main pedestrian flow by extracting the coordinate difference between the two endpoints of the centerline of the road boundary contour in the environmental spatial data; simultaneously, it reads the normal data of the frontal interaction surface pre-marked on the bounding box of the facility to be arranged, as the frontal orientation vector. This embodiment performs a dot product and inverse cosine operation on the three-dimensional spatial orientation vector and the frontal orientation vector to obtain the actual spatial angle value. Considering the floating-point precision overflow problem in the underlying calculation, this embodiment determines whether the actual spatial angle value falls within an interval centered at 90 degrees with an allowable deviation range of 2 degrees; when the determination result meets the requirements, a virtual connection edge representing a perpendicular orthogonal state is established between the corresponding security facility node and the main pedestrian flow node.

[0018] When handling line-of-sight interaction constraints, if the associated attribute includes clearance line-of-sight distance, this embodiment establishes virtual connection edges that satisfy the unobstructed line-of-sight association rules between the corresponding facility nodes. This embodiment uses the main stage node as the vertex of a cone, constructs spatial connections to the sound control console node, and uses this spatial connection as the central axis to generate a line-of-sight cone data model with a depth parameter of 120 meters and a preset cone angle parameter of 15 degrees. This embodiment traverses the obstacle coordinate matrix in the environmental spatial data to determine whether there are coordinate points falling inside the line-of-sight cone data model; after confirming no interference points, it generates virtual connection edges representing unobstructed line-of-sight. Finally, this embodiment transforms the node set containing all the facility nodes and the topological constraint edge set containing all the aforementioned virtual connection edges into an undirected weighted graph data structure and stores it in the system memory unit, thereby obtaining the spatial topological constraint model with computer readability and traversability, completing the underlying digital reconstruction of the layout rules.

[0019] Furthermore, the specific implementation process of steps 200-300 is as follows: This embodiment performs the step of initially assigning coordinates to the set of activity facilities within the three-dimensional feasible domain defined by the environmental spatial data to obtain a set of candidate layout schemes. This embodiment parses the previously acquired environmental spatial data to extract the physical entity boundary contours of the target site and the three-dimensional coordinate point set of unusable obstacle areas of existing building columns. After removing the coordinate space corresponding to the unusable obstacle areas from the overall site, this embodiment constructs a three-dimensional discrete grid network within the remaining three-dimensional feasible domain. The three-dimensional discrete grid network refers to a seamless array of cubic spatial units cut from continuous physical space according to a standard spatial resolution of 0.5 meters, serving as the minimum granularity coordinate reference for subsequent layout. Subsequently, this embodiment retrieves the total number of topological constraint edges connected to each activity facility in the spatial topological constraint model and extracts preset association weight values. The association weight values ​​are the normalized product coefficients of the facility's electrical hazard level index and the pedestrian density index. In this embodiment, the total number is multiplied by the associated weight value to obtain the composite topology evaluation score. The set of activity facilities is sorted in descending order according to the composite topology evaluation score. The top 15% of facilities are selected and defined as the core facility sequence with high connection density, and the remaining 85% of facilities are classified as the edge facility sequence.

[0020] This embodiment calculates the environmental attraction heat value of each grid node in the three-dimensional discrete grid network for each core activity facility in the core facility sequence, incorporating factors such as the distance to personnel evacuation exits and the location of load-bearing walls. The environmental attraction heat value is a scalar representing the spatial suitability of a specific grid node and the core activity facility under engineering safety conditions. Considering the presence of indoor obstacles, this embodiment does not use straight-line distance, but instead calls A... A heuristic pathfinding algorithm calculates the actual obstacle avoidance path length from the center point of each grid node to the nearest evacuation exit, extracts the reciprocal of this actual obstacle avoidance path length, and multiplies it by an evacuation weight coefficient set to 60%. Simultaneously, it calculates the three-dimensional Euclidean straight-line distance from the grid node to the nearest load-bearing wall and multiplies it by a load-bearing weight coefficient set to 40%. This embodiment performs a data processing step of summing the two weighted product results to obtain the environmental attraction heat value of the grid node. This embodiment performs a global traversal comparison of the entire network and forcibly binds the initial three-dimensional coordinates of the core activity facilities, prioritizing their allocation to the grid node with the highest environmental attraction heat value, thus completing the core coordinate allocation with deterministic location anchoring.

[0021] The expression for the environmental attraction thermal value is: ; in, The thermal value of environmental attractiveness; This is the evacuation weighting coefficient; This is the actual obstacle avoidance path length; This is the load-bearing weight coefficient; This represents the three-dimensional Euclidean linear distance. This embodiment uses the core activity facility that has completed the core coordinate allocation as the physical gravity center. Based on the topological constraint edges set in the spatial topological constraint model, it calculates the relative radiation region of each edge activity facility in the edge facility sequence. The relative radiation region refers to a closed three-dimensional spatial boundary that extends outward from the three-dimensional spatial coordinates of the core activity facility as the center and the maximum allowable assembly physical distance recorded by the topological constraint edge as the limit radius, and does not include the unusable obstacle region. Within the relative radiation region, this embodiment calls a pseudo-random number generator to perform random coordinate sampling on each edge activity facility to extract unoccupied three-dimensional discrete grid network nodes. When it is detected that 50 consecutive samples hit occupied nodes, this embodiment triggers an adaptive step-size expansion backoff mechanism, temporarily increasing the limit radius by 10% to break the deadlock state, until all edge facilities have completed position assignment, generating an overall facility coordinate state that includes the initial spatial positions of all facilities in a single arrangement.

[0022] To enrich the initial population diversity of the multi-objective optimization algorithm, this embodiment introduces a Gaussian perturbation operator to perform a total of 200 spatial mutation operations on the overall facility coordinate state, generating a candidate layout scheme set containing 200 spatial arrangement configurations. The Gaussian perturbation operator is a three-dimensional normal distribution translation transformation tool with the current coordinates of each facility as the expected mean and a perturbation deviation of 2.5 meters as the standard deviation. Subsequently, this embodiment executes the step of iteratively optimizing the candidate layout scheme set by calling the multi-objective optimization algorithm. This embodiment defines the underlying optimization direction: a minimum evacuation time evaluation function is constructed by calculating the effective evacuation time of the furthest personnel; a minimum wiring cost evaluation function is constructed by accumulating the actual wiring length of all physical cables between facilities; and a maximum view coverage evaluation function is constructed by calculating the total area of ​​the audience seats covered by the unobstructed view cone. This embodiment combines the above three independent orientation functions to form the multi-objective fitness evaluation system for multi-dimensional fitness scoring of any layout scheme.

[0023] This embodiment utilizes the multi-objective fitness evaluation system to perform iterative optimization on the candidate layout scheme set, cyclically performing non-dominated sorting based on crowding degree calculation and crossover mutation operations within an evolutionary generation cycle set to a maximum of 500 times. In each search cycle of the iterative optimization, this embodiment extracts in real-time the target activity facilities in the candidate layout scheme set that have undergone position changes. The trigger condition for the position change is that the absolute value of the change in the facility's three-dimensional absolute coordinates along any spatial axis is greater than a displacement sensitivity threshold set to 0.01 meters. This embodiment marks the activity facilities that have undergone changes as the target activity facilities with position changes in system memory, reads their current spatial three-dimensional coordinate data after the change, and retrieves their historical spatial three-dimensional coordinate data before the change. This embodiment calculates the spatial three-dimensional coordinate differences in three dimensions by subtracting the corresponding coordinate values ​​before the change from the changed horizontal, vertical, and vertical axis coordinate values, respectively. In this embodiment, the spatial three-dimensional coordinate differences of the three dimensions are merged in an orderly manner to serve as the coordinate offset vector representing the exact direction and distance of movement of the target facility. This vector is then written into a memory register to serve as the sole data source for triggering subsequent physical topology linkage updates.

[0024] Furthermore, the specific implementation process of steps 400-500 is as follows: This embodiment performs the step of locating the target facility node corresponding to the target activity facility based on the spatial topology constraint model. After obtaining the target activity facility that has undergone a location change, this embodiment extracts the unique device identification code assigned to the facility; the unique device identification code is a 64-bit hexadecimal string automatically generated by the system when entering the facility ledger. In this embodiment, in the node set of the spatial topology constraint model stored in the system memory, a hash mapping algorithm is invoked to convert the 64-bit hexadecimal string into the corresponding memory storage index value, thereby performing a non-traversal data hash matching retrieval to accurately locate the data structure that completely corresponds to the target activity facility, i.e., the target facility node. Subsequently, this embodiment traverses and retrieves all connecting edges in the topology constraint edge set that are associated with the target facility node through a physical memory address pointer. This embodiment identifies the other endpoint pointed to by each connecting edge in the undirected weighted graph structure, i.e., the subordinate node. To restore the physical spatial attributes of the layout scheme, this embodiment reverse-maps the three-dimensional coordinate parameters and equipment identifiers carried by the subordinate nodes to the original set of activity facilities, thereby extracting entity objects that have a specific engineering safety or physical distance dependency relationship with the target activity facility, i.e., the subordinate activity facilities. During this mapping process, this embodiment determines the specific connection edge with a unique weight attribute between the target facility node and the subordinate node as the target topological constraint edge representing the forced spatial coupling relationship between the two.

[0025] This embodiment performs the steps of reading the preset rigid constraint distance of the target topological constraint edge and calculating the initial topological deformation. This embodiment reads the static threshold parameter embedded within the underlying data structure of the target topological constraint edge and uses it as the rigid constraint distance; the static threshold parameter is a forced distance extreme value pre-imported from the National Activity Venue Fire Protection Planning Standard Database. The rigid constraint distance refers to the standard physical distance between two facilities while meeting fire protection and engineering compliance requirements; in the connection topology between the main stage and the sound control console, this value is set to 30 meters. Subsequently, this embodiment acquires the current absolute three-dimensional coordinates of the subordinate activity facility in real time, and the updated absolute three-dimensional coordinates of the target activity facility after generating the coordinate offset vector. This embodiment obtains the current spatial Euclidean distance between the two facilities by calculating the sum of the squares of the differences between the two sets of coordinates along the three orthogonal spatial axes and performing a square root operation. In this embodiment, the difference is calculated by subtracting the rigid constraint distance from the current spatial Euclidean distance. When the current spatial Euclidean distance is calculated to be 35 meters, the set rigid constraint distance of 30 meters is subtracted to produce a physical difference of 5 meters. This physical difference with positive and negative polarities is the initial topological deformation that characterizes the degree of destruction of the spatial topology.

[0026] This embodiment determines the sign of the initial topological deformation value to identify whether the target topological constraint edge is currently in a stretched or compressed state. When the initial topological deformation value is greater than 0 (positive sign), it indicates that the current facility spacing is excessively stretched, and this embodiment determines that the target topological constraint edge is in the stretched state. When the initial topological deformation value is less than 0 (negative sign), it indicates that the facility spacing is excessively compressed or even poses a risk of physical interference, and this embodiment determines that it is in the compressed state. Based on the identified stretched or compressed state, this embodiment matches the corresponding unequal-proportional nonlinear stiffness coefficient and damping attenuation coefficient in a preset feature parameter library. The preset feature parameter library is a two-dimensional relational matrix table stored in the system's read-only memory, recording engineering experience coefficient values ​​for different facility types and under different stress states. In this embodiment, for the tensile state, a nonlinear stiffness coefficient with a value of 1.5 and a damping attenuation coefficient with a value of 0.8 are matched from the two-dimensional relationship matrix table; for the compression state, considering the higher risk of physical interference, this embodiment matches a nonlinear stiffness coefficient with a value of 2.5 and a damping attenuation coefficient with a value of 0.9 to produce a stronger repulsion and recovery effect.

[0027] In this embodiment, the nonlinear stiffness coefficient, the damping attenuation coefficient, and the initial topological deformation are obtained through matching and used as underlying independent variables. These variables are input into a dedicated physics calculation thread created by the central processing unit, namely the preset elastic potential energy function, to calculate the constraint restoring force scalar used to converge the topological deviation. The internal data operation mechanism of the preset elastic potential energy function is as follows: First, the initial topological deformation is squared to construct a basic potential energy value. Then, this basic potential energy value is multiplied by the cube of the nonlinear stiffness coefficient to achieve a nonlinear constraint convergence effect where the larger the deformation deviation, the stronger the recovery penalty. Subsequently, in this embodiment, the above product result is multiplied again by the damping attenuation coefficient for smoothing and amplitude reduction processing. The final output scalar value is the constraint restoring force scalar.

[0028] The expression for the constraint restoring force scalar is: ; in, To constrain the restoring force scalar; The damping attenuation coefficient; It is a nonlinear stiffness coefficient; These are the initial topological deformations; In a specific layout optimization cycle, when the initial topological deformation under tension is 2 meters, the nonlinear stiffness coefficient is 1.5, and the damping attenuation coefficient is 0.8, this embodiment calculates the constraint restoration force scalar with a value of 10.8 by first calculating 2 squared to obtain 4, multiplying 4 by 1.5 cubed to obtain 13.5, and finally multiplying by 0.8. This scalar only represents the absolute numerical strength of topological restoration and does not include the directional attributes in three-dimensional space.

[0029] To convert the constraint restoring force scalar into the required three-dimensional spatial displacement, this embodiment extracts the spatial vector of the updated centroid coordinates of the target facility after its position change, pointed from the current centroid coordinates of the subordinate facility. This spatial vector is then divided by its geometric absolute magnitude to perform normalized data processing, resulting in a unit direction vector that represents only pure spatial pointing characteristics. This embodiment multiplies the previously calculated constraint restoring force scalar with the numerical components of the unit direction vector on the three orthogonal coordinate axes to complete the spatial projection mapping process, transforming the directionless scalar into a restoration compensation vector with three-dimensional pointing attributes. Finally, considering the initial inertia and motion trend of the target facility, this embodiment assigns an inertia inheritance weight of 0.3 to the coordinate offset vector previously written to the register. This embodiment multiplies the coordinate offset vector with its corresponding inertia inheritance weight of 0.3, and then performs a three-dimensional spatial vector addition operation with the restoration compensation vector to achieve weighted vector superposition. Through the mathematical synthesis operation that introduces inertial inheritance weights, this embodiment finally calculates the cooperative displacement vector that takes into account both topological distance following and global position stability, and sends it to the coordinate update driving module as the sole data basis for subsequent three-dimensional physical coordinate linkage correction of subordinate activity facilities.

[0030] The expression for the cooperative displacement vector is: ; in, For cooperative displacement vectors; Inertial inheritance weight; This is the coordinate offset vector; Constraint restoring force scalar; It is a unit direction vector; Furthermore, the specific implementation process of step 600 is as follows: This embodiment performs the step of superimposing the calculated cooperative displacement vector onto the current spatial coordinates of the subordinate activity facility to complete a single linked update of the spatial coordinates of the subordinate activity facility. This embodiment extracts the current absolute spatial coordinate values ​​of the subordinate activity facility in three dimensions stored in the dynamic random access memory, and retrieves the displacement component values ​​of the cooperative displacement vector calculated previously in the corresponding three spatial orthogonal axes. This embodiment uses an arithmetic logic unit to sum the two values ​​on the horizontal, vertical, and longitudinal axes respectively. To prevent physical boundary violations after displacement, this embodiment performs a feasible domain boundary value truncation check on the summation result. Specifically, this embodiment compares the summation result with the three-dimensional feasible domain boundary extreme values ​​recorded in the aforementioned environmental spatial data. If the horizontal axis summation result is 105.5 meters, and the maximum boundary extreme value of the site in that direction is 100 meters, this embodiment forcibly truncates the horizontal axis coordinate value to 100 meters. In this embodiment, the new three-dimensional coordinates after truncation verification are overwritten and written to the original storage address to achieve the single linkage update and ensure that facility movement is strictly restricted to legal physical space.

[0031] This embodiment uses the subordinate activity facility that has completed the single linkage update as the new target activity facility, and the cooperative displacement vector as the new coordinate offset vector, performing recursive tracing updates until there are no affected activity facilities in the spatial topology constraint model. This embodiment constructs a breadth-first search queue in memory; the breadth-first search queue is a first-in-first-out linear data cache structure. This embodiment pushes the node memory address of the subordinate activity facility that has completed coordinate correction into this queue, and marks the node with an access status boolean label of 1 to indicate that the node has been updated in the current iteration cycle, rejecting secondary enqueueing to prevent infinite loops caused by closed-loop topology. This embodiment cyclically performs dequeue operations, treating the popped node as the main body triggering the next layer of topological deformation, and again searching for the next-level subordinate node with a connected edge and an access status boolean label of 0 for displacement propagation. When the breadth-first search queue is completely empty, this embodiment determines that there are no more activity facilities affected by the linkage in the spatial topology constraint model, thus safely ending all the linkage updates in the current iteration cycle and completing the spatial translation self-consistency of the entire facility coupling community.

[0032] At the end of each iteration of optimization, this embodiment determines whether the iteration optimization has reached the preset convergence condition. First, it extracts the current frontier solution set output from the current iteration cycle and the historical frontier solution set output from the previous iteration cycle. In this embodiment, the frontier solution set refers to the group of optimal layout schemes at the first level after non-dominated sorting in multi-objective optimization. The data processing for non-dominated sorting involves pairwise comparison of the scores of all layout schemes in three dimensions: evacuation time, wiring cost, and view coverage. Layout schemes where no other scheme is superior to itself in all three dimensions are selected. This embodiment extracts a total of 50 elite layout schemes that meet this condition from the memory result matrix of the current search cycle to form the current frontier solution set. Simultaneously, this embodiment retrieves 50 elite schemes with the same data structure completely saved from the previous cycle from the cache register to form the historical frontier solution set, providing a time-series comparison benchmark for convergence determination.

[0033] This embodiment performs the step of calculating the solution distribution difference rate between the current frontier solution set and the historical frontier solution set. The solution distribution difference rate is a percentage indicator that measures the degree of drastic evolution of the overall population. This embodiment extracts the three-dimensional index scores of 50 schemes from the current frontier solution set and 50 schemes from the historical frontier solution set, and calculates the three-dimensional geometric centroid coordinates of each in the fitness space. This embodiment calculates the spatial Euclidean distance between these two three-dimensional geometric centroid coordinates and reads the pre-determined maximum theoretical diagonal length of the fitness space; this length is calculated by pre-extracting the multi-dimensional range distance between the three extreme points of the theoretical maximum evacuation time of the venue, the longest possible wiring distance, and 100% view coverage and the ideal zero point, and uses it as a unified benchmark. This embodiment divides the previously calculated spatial Euclidean distance by the maximum theoretical diagonal length of the fitness space, and finally multiplies it by 100 for normalization percentage conversion, thereby reducing the spatial offset of the high-dimensional solution set to an intuitive single value, which is used as the accurate solution distribution difference rate.

[0034] The expression for the dissimilarity rate of the solution set distribution is: ; in, The difference rate of the solution set distribution; The three-dimensional geometric centroid coordinates of the current frontier solution set are in the first... Numerical components of the fitness dimension; The three-dimensional geometric centroid coordinates of the solution set for the historical frontier are in the first... Numerical components of the fitness dimension; The maximum theoretical diagonal length value for the fitness space; In this embodiment, if the difference rate of the solution set distribution is less than the difference stability threshold for a consecutive preset number of times, the iterative optimization is determined to have reached the preset convergence condition, and the optimal layout scheme is obtained from the candidate layout scheme set that has completed the linkage update. This embodiment reads the difference stability threshold set to 0.5% and the preset number of times set to 15. When the difference rate of the solution set distribution calculated in a single iteration is less than 0.5%, the generation stability counter is incremented by 1; when the counter accumulates to 15, the population is determined to have lost its evolutionary activity and has reached the preset convergence condition. Subsequently, to eliminate the absolute numerical differences between evacuation time, wiring cost, and view area coverage caused by different dimensions, this embodiment first performs maximum-minimum range normalization on all indicators of all schemes within the frontier solution set, forcibly mapping all underlying indicators to a dimensionless decimal range of 0 to 1. Next, this embodiment reads the pre-configured engineering tendency weight coefficients, assigning a weight of 0.4 to the normalized evacuation time, a weight of 0.3 to the wiring cost, and a weight of 0.3 to the view area coverage. This embodiment multiplies each normalized decimal number by its corresponding weight and then sums the results to obtain the overall fitness of a single scalar. This embodiment selects the layout scheme with the highest overall fitness value and outputs the three-dimensional coordinate data of all facilities contained within it as the optimal layout scheme.

[0035] This embodiment also provides an automatic layout optimization system for activity facilities based on spatial topology constraints, including: The data acquisition and model building module is used to acquire environmental spatial data of the target site and a set of activity facilities to be arranged, and to build a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities. The initial coordinate allocation module is used to perform initial coordinate allocation on the set of activity facilities within the three-dimensional feasible domain defined by the environmental spatial data, so as to obtain a set of candidate layout schemes. The iterative optimization and offset recording module is used to call a multi-objective optimization algorithm to iteratively optimize the candidate layout scheme set, and during the iterative optimization process, extract the target activity facilities that have changed positions in the candidate layout scheme set in real time, and record the coordinate offset vector of the target activity facilities. The topology dependency extraction module is used to locate the target facility node corresponding to the target activity facility based on the spatial topology constraint model, and extract the subordinate activity facilities that have a physical dependency relationship with the target activity facility based on the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. The collaborative displacement calculation module is used to read the preset rigid constraint distance of the target topological constraint edge, and calculate the collaborative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector and the preset elastic potential energy function. The linkage update and scheme output module is used to perform linkage update on the spatial coordinates of the subordinate activity facilities based on the collaborative displacement vector until the iterative optimization reaches the preset convergence condition, and to obtain the optimal arrangement scheme from the candidate arrangement scheme set that has completed the linkage update.

[0036] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0037] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for automatic layout optimization of activity facilities based on spatial topology constraints, characterized in that, include: Obtain environmental spatial data of the target site and a set of activity facilities to be arranged, and construct a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities; Within the three-dimensional feasible domain defined by the environmental spatial data, the set of activity facilities is initially assigned coordinates to obtain a set of candidate layout schemes; A multi-objective optimization algorithm is invoked to iteratively optimize the candidate layout scheme set. During the iterative optimization process, the target activity facilities that have changed positions in the candidate layout scheme set are extracted in real time, and the coordinate offset vectors of the target activity facilities are recorded. Based on the spatial topology constraint model, the target facility node corresponding to the target activity facility is located, and the subordinate activity facilities that have a physical dependency relationship with the target activity facility are extracted according to the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. Read the preset rigid constraint distance of the target topological constraint edge, and calculate the cooperative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector, and the preset elastic potential energy function; The spatial coordinates of the subordinate activity facilities are updated in a coordinated manner based on the coordinated displacement vector until the iterative optimization reaches the preset convergence condition, and the optimal arrangement scheme is obtained from the set of candidate arrangement schemes that have completed the coordinated update.

2. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, The process of acquiring environmental spatial data of the target site and a set of activity facilities to be arranged, and constructing a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities, includes: Acquire the environmental spatial data of the target site and the set of activity facilities to be arranged; Extract the associated attributes of each activity facility in the set of activity facilities; Each activity facility in the set of activity facilities is abstracted into a corresponding facility node; Establish virtual connection edges between the facility nodes based on the associated attributes; The spatial topology constraint model is obtained by combining the set of nodes containing the facility nodes and the set of topology constraint edges containing the virtual connection edges.

3. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 2, characterized in that, The step of establishing virtual connection edges between the facility nodes based on the associated attributes includes: If the associated attribute includes a maximum cable connection threshold, then a virtual connection edge satisfying the physical ranging association rule is established between the corresponding facility nodes. If the associated attribute includes the normal angle of the main road, then a virtual connection edge satisfying the orientation alignment association rule is established between the corresponding facility nodes. If the associated attribute includes the clearance line of sight, then a virtual connection edge satisfying the unobstructed line of sight association rule is established between the corresponding facility nodes.

4. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, Within the three-dimensional feasible domain defined by the environmental spatial data, the initial coordinates of the set of activity facilities are assigned to obtain a set of candidate layout schemes, including: The environmental spatial data is analyzed to extract the boundary contours and unusable obstacle areas of the target site, and a three-dimensional discrete grid network is constructed within the three-dimensional feasible domain after removing the unusable obstacle areas. Based on the number and weight of the topological constraint edges connected to each activity facility in the spatial topological constraint model, the set of activity facilities is sorted in descending order to obtain the core facility sequence and the edge facility sequence. For each core activity facility in the core facility sequence, the environmental attraction thermal value of each grid node in the three-dimensional discrete grid network is calculated by integrating factors such as the distance to personnel evacuation exits and the location of load-bearing walls. The core activity facilities are preferentially assigned to the grid nodes with the highest environmental attraction thermal values ​​to complete the core coordinate assignment; Using the core activity facility that has completed the core coordinate allocation as the physical gravity center, and based on the topological constraint edges set in the spatial topological constraint model, calculate the relative radiation area of ​​each edge activity facility in the edge facility sequence; Random coordinate sampling within a constraint range is performed on each of the edge activity facilities within the relative radiation area to generate the overall facility coordinate state for a single initial coordinate allocation; A Gaussian perturbation operator is introduced to perform multiple spatial variation operations on the overall facility coordinate state to generate a set of candidate layout schemes containing multiple spatial arrangement configurations.

5. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, The multi-objective optimization algorithm is invoked to iteratively optimize the candidate layout scheme set. During the iterative optimization process, the target activity facilities that undergo position changes in the candidate layout scheme set are extracted in real time, and the coordinate offset vectors generated by the target activity facilities are recorded, including: Construct a multi-objective fitness evaluation system that includes evacuation time evaluation function, cabling cost evaluation function and view coverage evaluation function; The iterative optimization of the candidate arrangement scheme set is performed using the multi-objective fitness evaluation system. In each search cycle of the iterative optimization, the active facility whose coordinate position changes is marked as the target active facility that caused the position change; The spatial three-dimensional coordinate difference of the target activity facility before and after the change in coordinate position is calculated, and used as the coordinate offset vector generated by the target activity facility.

6. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, Based on the spatial topology constraint model, the process of locating the target facility node corresponding to the target activity facility and extracting subordinate activity facilities with physical dependencies on the target activity facility according to the set of topology constraint edges connected to the target facility node, thereby determining the target topology constraint edges between the target activity facility and the subordinate activity facilities, includes: In the node set of the spatial topology constraint model, retrieve the target facility node corresponding to the target activity facility; Traverse and retrieve all connecting edges in the set of topological constraint edges that are connected to the target facility node, and identify the subordinate node pointed to by each connecting edge; The subordinate nodes are reverse-mapped to the set of activity facilities to extract the subordinate activity facilities that have a physical dependency relationship with the target activity facility; The specific connection edge connecting the target facility node and the subordinate node is determined as the target topology constraint edge between the target activity facility and the subordinate activity facility.

7. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, The step of reading the preset rigid constraint distance of the target topological constraint edge, and calculating the cooperative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector, and the preset elastic potential energy function, includes: Real-time acquisition of the current spatial Euclidean distance between the subordinate activity facility and the target activity facility after the coordinate offset vector is generated; The initial topological deformation is obtained by calculating the difference between the current spatial Euclidean distance and the preset rigid constraint distance of the target topological constraint edge. Determine the sign of the initial topological deformation value to identify whether the target topological constraint edge is currently in a stretched or squeezed state; Based on the identified tensile or compressive state, a corresponding nonlinear stiffness coefficient and damping attenuation coefficient are matched in the preset elastic potential energy function. The nonlinear stiffness coefficient, the damping attenuation coefficient, and the initial topological deformation are input as independent variables into the preset elastic potential energy function to calculate the constraint restoring force scalar used to converge the topological deviation. Extract the unit direction vector from the subordinate activity facility to the target activity facility after the location change; The constraint restoring force scalar is spatially projected along the unit direction vector and then superimposed with the coordinate offset vector to calculate the cooperative displacement vector corresponding to the subordinate activity facility.

8. The method for automatic layout optimization of activity facilities based on spatial topology constraints according to claim 1, characterized in that, The process of performing a coordinated update on the spatial coordinates of the subordinate activity facilities based on the coordinated displacement vector until the iterative optimization reaches a preset convergence condition, and obtaining the optimal layout scheme from the set of candidate layout schemes that have completed the coordinated update, includes: The calculated collaborative displacement vector is superimposed on the current spatial coordinates of the subordinate activity facility to complete a single linkage update of the spatial coordinates of the subordinate activity facility; The subordinate activity facility that has completed the single linkage update is taken as the new target activity facility, and the collaborative displacement vector is taken as the new coordinate offset vector. Recursive tracing update is performed until there are no affected activity facilities in the spatial topology constraint model, and all linkage updates in the current iteration cycle are ended. At the end of each iteration of optimization, it is determined whether the iteration optimization has reached the preset convergence condition; If the preset convergence condition is met, the arrangement scheme with the highest comprehensive fitness is selected from the set of candidate arrangement schemes after all the linkage updates as the optimal arrangement scheme.

9. The automatic layout optimization method for activity facilities based on spatial topology constraints according to claim 8, characterized in that, The step of determining whether the iterative optimization has reached the preset convergence condition at the end of each iteration includes: Extract the current frontier solution set output in the current iteration cycle, and the historical frontier solution set output in the previous iteration cycle; Calculate the solution set distribution difference rate between the current frontier solution set and the historical frontier solution set; If the difference rate of the solution set distribution is less than the difference stability threshold for a consecutive preset number of times, then the iterative optimization is determined to have reached the preset convergence condition.

10. An automatic layout optimization system for activity facilities based on spatial topology constraints, characterized in that, include: The data acquisition and model building module is used to acquire environmental spatial data of the target site and a set of activity facilities to be arranged, and to build a spatial topology constraint model based on the association attributes of each activity facility in the set of activity facilities. The initial coordinate allocation module is used to perform initial coordinate allocation on the set of activity facilities within the three-dimensional feasible domain defined by the environmental spatial data, so as to obtain a set of candidate layout schemes. The iterative optimization and offset recording module is used to call a multi-objective optimization algorithm to iteratively optimize the candidate layout scheme set, and during the iterative optimization process, extract the target activity facilities that have changed positions in the candidate layout scheme set in real time, and record the coordinate offset vector of the target activity facilities. The topology dependency extraction module is used to locate the target facility node corresponding to the target activity facility based on the spatial topology constraint model, and extract the subordinate activity facilities that have a physical dependency relationship with the target activity facility based on the set of topology constraint edges connected to the target facility node, so as to determine the target topology constraint edges between the target activity facility and the subordinate activity facilities. The collaborative displacement calculation module is used to read the preset rigid constraint distance of the target topological constraint edge, and calculate the collaborative displacement vector of the subordinate activity facility based on the rigid constraint distance, the coordinate offset vector and the preset elastic potential energy function. The linkage update and scheme output module is used to perform linkage update on the spatial coordinates of the subordinate activity facilities based on the collaborative displacement vector until the iterative optimization reaches the preset convergence condition, and to obtain the optimal arrangement scheme from the candidate arrangement scheme set that has completed the linkage update.

Images