Public space decoration layout intelligent evaluation method based on safe evacuation

Abstract

This invention discloses an intelligent assessment method for the decoration layout of public spaces based on safe evacuation, belonging to the field of intelligent assessment technology. This intelligent assessment method for the decoration layout of public spaces based on safe evacuation first determines a baseline evacuation route, then simulates the process of people prioritizing movement along the baseline route, autonomously optimizing their path after encountering obstruction, and actively returning to the baseline route. This reproduces the path decision-making logic and detour behavior characteristics of people in evacuation scenarios. Simultaneously, by quantitatively constructing a passage resistance coefficient through detour path parameters, personnel behavior data during the path optimization stage, and data on the occupancy of decoration components in the passageway from multiple dimensions, it achieves a precise and quantifiable assessment of the impact of decoration layout on evacuation passage, rather than traditional static compliance verification. This realizes a shift from passive compliance assessment to proactive layout optimization, ensuring that the assessment results of the decoration layout closely match the actual evacuation behavior patterns of people.

Description

Technical Field

[0001] This invention relates to the field of intelligent assessment technology, specifically to an intelligent assessment method for the decoration and layout of public spaces based on safe evacuation. Background Technology

[0002] In the scenario of safety evacuation assessment of public space decoration layout, existing technologies have the following shortcomings: First, existing technologies only conduct evacuation assessments based on static geometric parameters and do not construct a benchmark evacuation channel system that conforms to the actual behavioral habits of people during emergency evacuations. Second, existing technologies cannot simulate the dynamic detour behavior and path optimization decision-making process of people prioritizing conventional evacuation routes after decoration components are installed and then deviating from the original channels when encountering obstructions; they can only complete static compliance verification. Third, existing technologies have not established a multi-dimensional quantitative system for the impact of passage, resulting in a deviation between the assessment results and the evacuation efficiency in real-world scenarios, and failing to achieve intelligent optimization of decoration layout parameters with the goal of minimizing evacuation passage resistance.

[0003] Therefore, there is an urgent need for an intelligent evaluation method for the decoration layout of public spaces based on safe evacuation, which can be based on the dynamic detour and path optimization decision-making process of simulating personnel encountering obstacles and realize intelligent optimization of decoration layout parameters. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent evaluation method for the decoration layout of public spaces based on safe evacuation. This method solves the problem that existing technologies evaluate decoration layouts based solely on static geometric parameters, which cannot simulate dynamic detour behavior and path decision-making processes of people, leading to discrepancies between the evaluation results and the actual evacuation efficiency.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a smart evaluation method for the decoration layout of public spaces based on safe evacuation, comprising the following steps: determining the benchmark evacuation route based on the movement trajectory of a simulated person in a public space model.

[0006] After loading the planned decoration components into the public space model, the simulated people are restricted to prioritize moving along the baseline evacuation route, and abnormal nodes are identified based on the movement data.

[0007] If the simulated human in the area in front of the abnormal node is allowed to leave the baseline evacuation route to find a path, while the simulated human is restricted from trying to return to the baseline evacuation route by the shortest path, the regression point is determined.

[0008] The traffic resistance coefficient of abnormal nodes is determined by the detour path parameters between abnormal nodes and regression points, the data from the path optimization stage, and the occupancy data of planned decoration components on the baseline evacuation route.

[0009] Parameter optimization conditions are constructed to optimize the parameter set of the planned decoration components with the goal of minimizing the traffic resistance coefficient.

[0010] Compared with existing technologies, this invention has the following beneficial effects: By first determining the baseline evacuation route, and then simulating the process of personnel prioritizing movement along the baseline route, autonomously optimizing their path after encountering obstruction, and actively returning to the baseline route, this invention reproduces the path decision-making logic and detour behavior characteristics of personnel in evacuation scenarios. At the same time, by constructing a passage resistance coefficient through multi-dimensional quantification using detour path parameters, personnel behavior data during the path optimization stage, and data on the occupancy of passages by decoration components, this invention achieves a precise and quantifiable assessment of the impact of decoration layout on evacuation passage, rather than traditional static compliance verification. This realizes a shift from passive compliance assessment to proactive layout optimization, ensuring that the assessment results of decoration layout align with the actual evacuation behavior patterns of personnel. Attached Figure Description

[0011] Figure 1 This is a flowchart of the intelligent assessment method for the decoration and layout of public spaces based on safe evacuation, as proposed in this invention.

[0012] Figure 2 This is a flowchart for determining the baseline evacuation route in the intelligent evaluation method for public space decoration layout based on safe evacuation of the present invention.

[0013] Figure 3 This is a flowchart illustrating the identification of abnormal nodes in the intelligent evaluation method for the decoration and layout of public spaces based on safe evacuation, as described in this invention.

[0014] Figure 4 This is a flowchart for determining the regression point in the intelligent evaluation method for public space decoration layout based on safe evacuation in this invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Please refer to the accompanying drawings. Figure 1 The present invention provides a technical solution: a method for intelligent evaluation of public space decoration layout based on safe evacuation, including the following steps: S1, determining the benchmark evacuation channel based on the movement trajectory of a simulated person in a public space model.

[0016] To objectively identify naturally occurring, habitual evacuation routes in public spaces from the perspective of space users, rather than relying on geometric centerlines on design drawings or artificially defined evacuation routes, by simulating the autonomous exploration behavior of a large number of individuals under unconstrained conditions, this approach transforms the determination of evacuation routes from static spatial measurements to dynamic behavioral simulation. This allows benchmark routes to realistically reflect the most likely movement trajectories people would choose in an emergency, such as... Figure 2As shown, the specific process is as follows: S101, Discretize the horizontal projection of the public space model into a grid, and assign a basic access cost to each grid. The grid side length is set to 0.5 meters based on the average shoulder width of a human. The basic access cost is 1, representing the unit time required for a person to pass through the grid under barrier-free conditions.

[0017] S102. Set the starting point and ending point of the evacuation, and deploy the simulators. Use the sum of the basic travel costs traversed by the simulators from the starting point to the current grid along the current path as the actual travel cost. Use the Euclidean distance from the current grid to the ending point as the estimated travel cost.

[0018] S103. The sum of the actual passage cost and the estimated passage cost is used as the estimated total cost from the current grid to all adjacent passable grids.

[0019] S104. In each step of the path expansion of each simulated person, when selecting the next passable grid from the current grid, a random perturbation term is applied to the estimated total cost as the travel cost, and the passable grid with the minimum travel cost is selected as the target for the next move. The random perturbation term is the value of a random number uniformly distributed in the interval [-0.5, 0.5] meters multiplied by the grid side length.

[0020] S105. Record the grids that the simulated person passes through in sequence as a motion trajectory.

[0021] S106. A density-based spatial clustering algorithm is used to cluster motion trajectories, obtaining the top K clusters containing the most motion trajectories. The clustering radius is set to the average shoulder width of a human body of 0.5 meters, and the minimum number of cluster points is set to the product of 0.05 and the number of simulated people. K is the preset number of baseline channels, ranging from 1 to 5.

[0022] S107. For each selected cluster, align the trajectory points of all motion trajectories within the cluster, calculate the average coordinates, and obtain the center trajectory.

[0023] S108. Based on the center trajectory, extend the distance to both sides by 0.75 meters (equivalent to 1.5 times the width of a human shoulder) to obtain the baseline evacuation channel.

[0024] This embodiment uses the A* algorithm to ensure that the path of each simulated person is rational in the overall direction (moving towards the destination), while simulating the differences in individual decisions through random perturbation; the spatial clustering algorithm can identify the spatial clustering characteristics of the trajectory, eliminate random abnormal trajectories, and finally obtain the representative benchmark channel adopted by the majority of people, avoiding the defect that a single shortest path cannot reflect the diversity of actual paths.

[0025] The base passage cost represents the baseline time cost required for a simulated person to travel one unit distance on an unobstructed grid; its value is merely a counting unit. The actual passage cost reflects the length cost of the path. The estimated passage cost reflects the ideal length cost of the remaining path. The sum of the two, the estimated total cost, is used to guide the simulated person to always move towards the destination. All cost values ​​are based on grid edge length to ensure that path selection comparisons are conducted within the same dimensional system.

[0026] In addition, the purpose of applying a random perturbation term is to break the deterministic path selection result of the A* algorithm, so that the trajectory of each simulated person produces individual differences that are consistent with reality. By applying random fluctuations to the estimated total cost at each decision point, the simulated person can produce a diverse trajectory distribution while maintaining the general direction of moving towards the destination.

[0027] It should be noted that public space models can be constructed in professional simulation platforms based on intelligent agent modeling by importing building floor plan data. For example, the evacuation simulation software Pathfinder can realize geometric modeling, grid generation, path planning, and behavior simulation of building space. At the same time, the GAMA open-source simulation platform supports the integration of Shapefile spatial data to build three-dimensional building models and perform evacuation simulations. The path optimization in the entire simulation process can be implemented in the Python programming environment using the A* algorithm. By discretizing the space into grids and assigning a passage cost to each grid, the path planning and trajectory generation from the starting point to the destination are completed.

[0028] S2. After loading the planned decoration components into the public space model, restrict the simulated people to prioritize moving along the baseline evacuation route, and determine abnormal nodes based on the movement data.

[0029] To quantify the differences in traffic flow in baseline evacuation routes before and after the installation of planned renovation components using statistical methods, and to identify locations where traffic efficiency is reduced due to the occupancy of these components, such as... Figure 3 As shown, the specific process for identifying abnormal nodes is as follows: S201, sampling points are set at equal intervals along the centerline of each baseline evacuation route at a predetermined spacing. For example, the spacing is set to 0.5m to ensure that the spatial resolution can capture the details of changes in the passage status of personnel.

[0030] S202. For each simulator, if the vertical distance between its current position and the centerline is greater than a set distance threshold, a correction velocity pointing towards the centerline is applied to bring the simulator back to the centerline. For example, the set distance threshold is 0.5m.

[0031] S203. For each sampling point, if the average speed of the corresponding simulated person is less than a set speed threshold and the average density is greater than a set density threshold, it is marked as a candidate node. The set speed threshold is the measured average speed of the sampling point without the planned decoration components loaded minus twice the speed standard deviation, and the set density threshold is the measured average density of the sampling point without the planned decoration components loaded plus twice the density standard deviation.

[0032] S204. Among all candidate nodes, the sampling point closest to the starting point along the direction of the baseline evacuation channel is identified as an abnormal node.

[0033] The simulated average speed is the average of the instantaneous speeds of simulated people at the corresponding sampling points. The average density is the average of the instantaneous population density within a 1-meter radius of the corresponding sampling point. The measured average speed and measured average density are also obtained using the same calculation method described above.

[0034] In this embodiment, abnormal nodes are defined as locations where the speed decreases and the density increases compared to the baseline state without decorations. This allows the identification of abnormal nodes to be based on repeatable and verifiable data comparisons. At the same time, the rule of identifying the abnormal candidate point closest to the starting point ensures that the abnormal location encountered by the personnel is identified first.

[0035] S3. If the simulated human in the area in front of the abnormal node is allowed to leave the baseline evacuation route to find the path, while the simulated human is restricted from trying to return to the baseline evacuation route by the shortest path, the return point is determined.

[0036] The process of determining whether a simulated person within a designated area in front of the abnormal node is allowed to leave the baseline evacuation route for path optimization is as follows: S301, using the location of the abnormal node as the rear boundary, extend a preset distance in the opposite direction of the baseline evacuation route as the set observation range. For example, the preset distance is set to 5 meters to ensure sufficient space to assess the situation ahead before approaching the abnormal point. The observation range is a strip-shaped area extending along the centerline of the baseline evacuation route, with a width equal to the width of the influence area of ​​the baseline evacuation route.

[0037] S302. If the average speed of the simulated human entering the observation range is less than the set speed threshold and the duration exceeds the set time, or the average density is greater than the set density threshold and the duration exceeds the set time, then path optimization is allowed.

[0038] For example, the speed threshold is set to 0.5 m / s and the density threshold is set to 2 people / square meter.

[0039] like Figure 4As shown, the process of determining the regression point is as follows: S303, take the current position of the simulated person as the starting point for optimization, and take the first area behind the abnormal node that is not affected by the planned decoration as the target area. The target area is the first area behind the abnormal node (away from the starting point) that is not affected by the planned decoration. It is a continuous space within the influence area of ​​the baseline evacuation passage, and there are no decorations occupying it.

[0040] S304. Take all grids in the target area as candidate target points, and select the grid with the smallest Euclidean distance from the optimization starting point as the virtual target point.

[0041] S305. For any grid between the optimization starting point and the virtual target point, the corresponding actual travel cost is the sum of the basic travel costs of the grids traversed from the optimization starting point to the current grid. The estimated travel cost is the Euclidean distance from the current grid to the virtual target point. The regression preference term is the product of the shortest distance from the current grid to the boundary of the influence area of ​​the baseline evacuation channel and the preference coefficient.

[0042] S306. With the goal of minimizing the sum of actual travel cost, estimated travel cost, and regression preference term, search for the optimal path from the optimization starting point to the virtual target point, and use this path as the optimization path.

[0043] S307. The simulator moves along the optimization path. The position where the simulator first enters the influence area of ​​the baseline evacuation channel is marked as the regression point, and the optimization is terminated. After that, the simulator continues to move along the baseline evacuation channel.

[0044] S308. If the simulated person discovers during the movement that the virtual target point can be reached directly from the current position along the benchmark evacuation route without detour, the optimization will be terminated in advance, and the person will move directly along the benchmark evacuation route. The first position in the influence area of ​​the benchmark evacuation route will be marked as the regression point.

[0045] This embodiment uses A* algorithm plus regression preference to determine the regression point. This ensures that the detour path avoids construction debris while guiding the simulated person to prioritize routes closer to the baseline passage. This aligns with the psychological expectation that people tend to return to familiar paths as quickly as possible. By penalizing grids far from the passage through the regression preference term, it avoids the irrational paths that might result from the simple A* algorithm, which excessively deviate from the original passage in pursuit of the shortest distance. This ensures that the optimization result simultaneously satisfies obstacle avoidance requirements and regression intentions, more closely reflecting actual evacuation behavior.

[0046] The purpose of introducing a regression preference term is to embed the behavioral constraint of reverting to the original path into the path optimization algorithm, preventing the simulated person from continuously deviating from the baseline path and entering unpredictable areas after detouring. By quantifying the distance between the current grid and the influence area of ​​the baseline path, the path selection, while meeting obstacle avoidance requirements, always possesses an attraction pointing towards the path, ensuring that the detour path has a clear regression intention. This makes the determination of the regression point logically necessary, rather than a result of random exploration.

[0047] The preference coefficient is a negative reciprocal of the grid side length, ensuring that the value of the regression preference term is on the same order of magnitude as the passage cost. The penalty increases with distance from the passage, thus guiding the path naturally closer to the passage. The preference coefficient ranges from -0.5 / m to -2 / m. In this embodiment, the grid side length is 0.5m, so -2 / m is used. When there are many obstacles around the benchmark passage in the public space, and it is necessary to allow the simulated person to detour over a larger area, the preference coefficient can be adjusted to -0.5 / m to -1 / m to reduce the regression constraint strength.

[0048] S4. Determine the traffic resistance coefficient of the abnormal node using the detour path parameters between the abnormal node and the return point, the path optimization stage data, and the occupancy data of the planned decoration components on the benchmark evacuation channel.

[0049] The detour path parameters include the actual length and average curvature of the detour path. Path optimization phase data includes the angle between the simulated person's direction of travel at each step and the tangent to the centerline of the baseline evacuation route, as well as the simulated person's average speed within the initial exploration segment. Occupancy data includes the width and length of the fixture intrusion.

[0050] The initial exploration segment is 25% of the total length of the detour path. This length can cover the entire behavioral phase of the simulated person after leaving the baseline channel, from decision-making hesitation to determining the detour direction. For detour paths with a total length of less than 5 meters, the initial exploration segment is uniformly taken as the first 1 meter.

[0051] It should be noted that the data from the optimization phase reflects the degree of movement disorder caused by the simulator avoiding other simulators. The occupancy data reflects the severity of compression of the physical space in the passageway.

[0052] When no simulated pedestrians are using the baseline evacuation route, the resistance mainly comes from the physical obstruction of the route by the decorative elements and the geometric characteristics of the detour path. When simulated pedestrians are using the route, the dynamic impact of other simulated pedestrians traveling normally along the baseline route on the detour pedestrians needs to be considered. Therefore, it is necessary to determine the passage resistance coefficient of abnormal nodes in different cases. The process is as follows: S401, calculate the ratio of the actual length of the detour path to the straight-line distance from the abnormal node to the return point, and use this ratio as the path extension factor. The magnitude of the path extension factor directly reflects the increased distance cost caused by forced detours due to obstruction by decorative elements; a larger value indicates a longer detour distance and more serious space waste.

[0053] S402. Discretize the detour path into multiple points, calculate the average curvature of the path based on the local curvature determined by three adjacent points, and use the product of the average curvature and the actual length of the detour path as the path curvature factor.

[0054] S403. The path extension factor and the path curvature factor are linearly weighted and summed to obtain the geometric resistance of the bypass path.

[0055] S404. If there are simulated people traveling along the baseline evacuation route in the section behind the abnormal node, the passage resistance coefficient shall be determined by the geometric resistance of the detour path and the data from the path optimization stage.

[0056] S405. If there are no simulated people traveling along the baseline evacuation route in the section behind the abnormal node, the passage resistance coefficient shall be determined by the geometric resistance of the detour path and the occupancy data.

[0057] The path extension factor directly reflects the increased distance cost caused by detours due to obstructions from decorative elements; a larger value indicates a longer detour and greater space waste. Average curvature reflects the degree of curvature per unit length of the path; multiplying it by the path length yields the cumulative curvature of the entire detour path. The path curvature factor is equivalent to the sum of directional changes at all points on the path (in radians); a larger value indicates a more tortuous path with sharper or more frequent turns. The geometric resistance of the detour path reflects the degree of spatial obstruction caused by the decorative elements; it is independent of the presence of other simulated individuals and represents the alteration of the path's inherent properties by the decorative elements themselves.

[0058] In this embodiment, the weight coefficient corresponding to the path extension factor is 0.7, and the weight coefficient corresponding to the path curvature factor is 0.3. This weight ratio is determined based on engineering experience that the influence of path length on evacuation time is greater than that of path curvature in evacuation scenarios. In densely populated scenarios, the second weight coefficient can be appropriately increased to 0.4 and the first weight coefficient can be decreased to 0.6 to adapt to the greater impact of winding paths on people's movement.

[0059] In order to quantify the chaotic costs caused by decision-making hesitation, directional probing, and speed adjustments by simulating the behavior of a human as they leave the baseline evacuation route and begin to search for a new path in the initial exploration phase, the process of determining the traffic resistance coefficient using data from the geometric resistance of the detour path and the path optimization phase is as follows: the first section of the detour path is designated as the initial exploration phase, and the change in direction angle is calculated based on the angle between the direction of the human's movement at each step in the initial exploration phase and the tangent direction of the centerline of the baseline evacuation route, which is used as the direction swing factor.

[0060] The ratio of the average speed when there are no decorations on the baseline channel to the average speed of the simulated person in the initial exploration section is used as the speed reduction factor.

[0061] The chaotic drag of the bypass path is obtained by linearly weighting and summing the directional sway factor and the velocity reduction factor.

[0062] The traffic resistance coefficient is obtained by adding the geometric resistance of the detour path to the chaotic resistance of the detour path.

[0063] The formula for calculating the directional oscillation factor is as follows: ,in, The directional swing factor, For step index, This represents the total number of steps within the initial exploration phase. It is the first The angle between the direction of movement and the centerline of the reference evacuation route. It is the sum of the absolute values ​​of the changes in direction angle between all adjacent steps within the entire initial exploration phase, ultimately yielding the average direction change amplitude for each step, also known as the direction sway factor.

[0064] The directional sway factor reflects the degree of directional stability of the simulated human when initially searching for a new path after leaving the baseline channel. A higher value indicates more drastic directional changes and a more unpredictable route, suggesting the simulated human lacks a clear judgment of its detour direction. The speed reduction factor reflects the degree of deceleration of the simulated human in the initial stage of detour. A value greater than 1 indicates a speed reduction; a higher value indicates a more severe deceleration and a greater loss of travel efficiency.

[0065] The obstacle course confusion reflects the behavioral costs incurred by the simulated person in the initial stages of detour due to decision-making confusion. A higher value indicates that the simulated person needs to expend more cognitive resources and time processing direction selection and speed adjustments, reflecting the degree to which the decorations disrupt the smoothness of the person's behavior.

[0066] The third weighting coefficient for the directional sway factor is set to 0.4, and the fourth weighting coefficient for the speed decrease factor is set to 0.6. This weighting ratio is determined based on the fact that the impact of speed loss on traffic efficiency is greater than that of directional sway in evacuation scenarios. In scenarios with low personnel density, the fourth weighting coefficient can be appropriately reduced to 0.5 and the third weighting coefficient can be increased to 0.5.

[0067] In this embodiment, the directional sway factor reflects the degree of confusion in route selection, and the speed reduction factor reflects the degree of loss of travel efficiency, which can quantify the decision-making burden at the initial stage of detour. This allows the traffic resistance coefficient to simultaneously include the costs of spatial obstacles and behavioral disorder, avoiding the difficulty in accurately reflecting actual evacuation by using only path length or curvature.

[0068] The process of determining the traffic resistance coefficient based on the geometric resistance and occupancy data of the detour path is as follows: calculate the ratio of the intrusion width of the decoration component to the original width of the reference passage, and use it as the width occupancy factor.

[0069] Calculate the ratio of the intrusion length of the decoration component to the actual length of the detour path, and use it as the length occupancy factor.

[0070] Multiplying the width occupancy factor by the length occupancy factor yields the resistance affected by channel occupancy.

[0071] The traffic resistance coefficient is obtained by adding the geometric resistance of the detour path to the resistance caused by the occupancy of the passage.

[0072] The width occupancy factor represents the proportion of lateral passage space that is compressed. The closer the value is to 0, the less severe the lateral encroachment; the closer it is to 1, the more severe the lateral encroachment, even to the point of complete blockage. The length occupancy factor represents the proportion of the section encroached by decorations in the entire detour path. The closer the value is to 0, the smaller the encroached section is in the path; the closer it is to 1, the more the entire detour path is within the encroachment area.

[0073] The product of the width occupancy factor and the length occupancy factor reflects the degree to which the decorative elements encroach on the baseline passageway in a two-dimensional plane. Multiplying two values ​​less than 1 further reduces the result, physically representing the proportion of space occupied. The larger the value, the more comprehensive the physical deprivation of the passageway by the decorative elements, and the more severe the spatial obstruction caused by the decorative elements for evacuation.

[0074] In this embodiment, the capacity loss of a single-point section is quantified by the width occupancy factor, and the proportion of the affected section in the entire detour path is quantified by the length occupancy factor. The two are multiplied to obtain the comprehensive occupancy degree in two-dimensional space, so that the drag coefficient is based on the measurement of the change in the geometric properties of the channel, and the drag coefficient reflects the degree of physical space deprivation by the decoration components in the case of no one behind.

[0075] S5. Construct parameter optimization conditions to optimize the parameter set of the planned decoration components with the goal of minimizing the traffic resistance coefficient.

[0076] The parameter set for the planned decoration components includes width, depth, and center coordinates. Considering that multiple parameters of the planned decoration components can be adjusted simultaneously, the number of parameter combinations increases geometrically with the dimension, making global traversal computationally infeasible. Therefore, it is necessary to adopt a dimension-by-dimensional expansion and adjustment method. Specifically: S501, the parameter optimization condition is that the sum of the occupancy width adjustable range, occupancy depth adjustable range, installation center coordinate adjustable range, and the occupancy width of all planned decoration components within the same slice is not greater than the difference between the reference net width of the passage and the minimum evacuation net width specified in the fire protection code.

[0077] The reference net width of the passage refers to the original passage width within the area affected by the reference evacuation passage when no planned decorations are loaded. It is the lateral width of the strip area formed after extending 0.75 meters to both sides of the central trajectory in step S108. An exemplary value of 1.5 meters can be taken.

[0078] S502. Determine the priority of each planned decoration component.

[0079] S503. For the first planned decoration component to be optimized, within the adjustable range of the occupied width, adjust the step size according to the preset parameters to generate alternative values ​​for the occupied width, forming a set of feasible domains for the occupied width.

[0080] S504. Traverse all candidate values ​​for occupied width within the feasible region set for occupied width, and simultaneously generate candidate values ​​for occupied depth by adjusting the step size according to preset parameters within the adjustable range for occupied depth, thus forming a two-dimensional feasible region set for parameters.

[0081] S505. Traverse all parameter combinations within the feasible domain of the two-dimensional parameters, and simultaneously generate alternative values ​​for the installation center coordinates by adjusting the step size according to the preset parameters within the adjustable range of the installation center coordinates, thus forming a three-dimensional parameter combination.

[0082] S506. For each combination of three-dimensional parameters, calculate the corresponding traffic resistance coefficient.

[0083] S507. Use the combination of three-dimensional parameters corresponding to the minimum traffic resistance coefficient as the optimized parameters of the current planned decoration component to be optimized.

[0084] S508. Continue to optimize the next priority planning decoration components.

[0085] In this embodiment, priority sorting ensures that the decorative components that contribute the most to traffic resistance receive the best optimization resources first. By constructing the feasible domain of parameters layer by layer, it ensures that each search expands on the optimal solution of the previous dimension, avoiding the computational explosion caused by traversing all parameter combinations at the same time, and achieving a balance between computational efficiency and optimization accuracy in the optimization process.

[0086] The segment between the abnormal node and the regression point is sliced ​​according to a set interval to obtain several slices. The process of determining the priority of each planned decoration component is as follows: if there are multiple planned decoration components on a certain slice, the width occupancy factor of each planned decoration component at the slice is calculated. The width occupancy factor represents the degree of occupancy impact of a single channel, and the priority is determined according to the degree of occupancy impact of a single channel from large to small.

[0087] If there are multiple planned decoration components, but no multiple planned decoration components exist on any slice, then the priority is determined according to the distance between the planned decoration component and the starting point of the evacuation route, from largest to smallest, with the larger the distance, the higher the priority.

[0088] If there is only one planned decoration component, then it is directly identified as the object to be optimized.

[0089] In this embodiment, when multiple decorative elements exist on a slice, the element with the largest width occupancy factor is optimized first, achieving the greatest reduction in traffic resistance with minimal adjustment cost. When there are no multiple decorative elements on a slice but multiple elements exist within a segment, the decorative element furthest from the starting point is optimized first, releasing traffic pressure in the subsequent space in advance and preventing previous optimizations from being disrupted by subsequent elements. When there is only one decorative element, it is directly used as the optimization target, eliminating the need for sorting and wasting computational resources. This hierarchical processing ensures that each optimization iteration generates the maximum marginal benefit.

[0090] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0091] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0092] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0093] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0094] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent assessment of public space decoration layout based on safe evacuation, characterized in that, Includes the following steps: Baseline evacuation routes are determined based on the simulated movement trajectories of people in a public space model. After loading the planned decoration components into the public space model, the simulated people are restricted to prioritize moving along the baseline evacuation route, and abnormal nodes are identified based on the movement data; If the simulated human in the area in front of the abnormal node is allowed to leave the baseline evacuation route to find the path, while the simulated human is restricted from trying to return to the baseline evacuation route by the shortest path, the regression point is determined. The passage resistance coefficient of the abnormal node is determined by the detour path parameters between the abnormal node and the regression point, the data of the path optimization stage, and the occupancy data of the planned decoration components on the benchmark evacuation channel. Parameter optimization conditions are constructed to optimize the parameter set of the planned decoration components with the goal of minimizing the traffic resistance coefficient.

2. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The process of determining the baseline evacuation route based on the simulated human movement trajectory in a public space model is as follows: Discretize the horizontal projection of the public space model into a grid and assign a basic passage cost to each grid; Set the starting point and ending point of the evacuation, and use the sum of the basic travel costs traversed from the starting point to the current grid along the current path as the actual travel cost; Use the Euclidean distance from the current grid to the destination as the estimated passage cost; The sum of the actual passage cost and the estimated passage cost is used as the estimated total cost from the current grid to all adjacent passable grids; In each step of the path expansion of each simulated person, when selecting the next traversable grid from the current grid, a random perturbation term is applied to the estimated total cost as the travel cost, and the traversable grid with the minimum travel cost is selected as the target for the next move. Record the grids that the simulated person passes through in sequence as a motion trajectory; Density-based spatial clustering algorithms are used to cluster motion trajectories to obtain the top K clusters containing the most motion trajectories. For each selected cluster, align the trajectory points of all motion trajectories within the cluster, calculate the average coordinates, and obtain the center trajectory. Using the central trajectory as a reference, extend a set distance to both sides to obtain the reference evacuation channel.

3. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The process of restricting simulated humans to prioritize traveling along the baseline evacuation route and determining abnormal nodes based on travel data is as follows: Sampling points are set at equal intervals along the centerline of each reference evacuation route at a predetermined spacing; For each simulator, if the vertical distance between the current position and the center line is greater than a set distance threshold, a correction velocity pointing towards the center line is applied to bring the simulator back to the center line. For each sampling point, if the average speed of the corresponding simulated person is less than a set speed threshold and the average density is greater than a set density threshold, it is marked as a candidate node. The sampling point closest to the starting point along the reference evacuation route among all candidate nodes is identified as an abnormal node.

4. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The process of determining whether to allow simulated humans within a designated area in front of an abnormal node to leave the baseline evacuation route for path optimization is as follows: The observation range is defined by extending a preset distance in the opposite direction of the baseline evacuation channel, with the location of the abnormal node as the rear boundary. If the average speed of the simulated human entering the observation range is less than the set speed threshold and the duration exceeds the set time, or the average density is greater than the set density threshold and the duration exceeds the set time, then path optimization is allowed.

5. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The simulated human leaves the baseline evacuation route and performs path optimization, while simultaneously being restricted to attempting to return to the baseline evacuation route via the shortest path. The process of determining the return point is as follows: The current position of the simulator is taken as the starting point for optimization, and the first area behind the abnormal node that is not affected by the planned decoration component is taken as the target area. All grids within the target area are used as candidate target points, and the grid with the smallest Euclidean distance from the optimization starting point is selected as the virtual target point. For any grid between the optimization starting point and the virtual target point, the corresponding actual passage cost is the sum of the basic passage costs of the grids traversed from the optimization starting point to the current grid, the estimated passage cost is the Euclidean distance from the current grid to the virtual target point, and the regression preference term is the product of the shortest distance from the current grid to the boundary of the influence area of ​​the baseline evacuation channel and the preference coefficient. With the goal of minimizing the sum of actual travel cost, estimated travel cost, and regression preference term, the optimal path from the optimization starting point to the virtual target point is searched as the optimization path; The simulator moves along the optimization path, and the position where the simulator first enters the influence area of ​​the baseline evacuation route is marked as the regression point, and the optimization is terminated. After that, the simulator continues to move along the baseline evacuation route.

6. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The process of determining the traffic resistance coefficient of abnormal nodes is as follows: Calculate the ratio of the actual length of the detour path to the straight-line distance from the abnormal node to the return point, and use this ratio as the path extension factor; The detour path is discretized into multiple points. The average curvature of the path is calculated based on the local curvature determined by three adjacent points. The product of the average curvature and the actual length of the detour path is used as the path curvature factor. The geometric resistance of the bypass path is obtained by linearly weighting and summing the path extension factor and the path curvature factor. If there are simulated people traveling along the baseline evacuation route in the section behind the abnormal node, the passage resistance coefficient is determined by the geometric resistance of the detour path and the data from the path optimization stage. If there are no simulated people traveling along the baseline evacuation route in the section behind the abnormal node, the passage resistance coefficient is determined by the geometric resistance of the detour path and the occupancy data.

7. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 6, characterized in that, The process of determining the traffic resistance coefficient using data from the detour route geometric resistance and the route optimization stage is as follows: The first section of the detour path is designated as the initial exploration section. The change in the direction angle is calculated based on the angle between the direction of each step of the simulated person in the initial exploration section and the tangent direction of the center line of the reference evacuation channel, and is used as the direction swing factor. The ratio of the average speed when there are no decorations on the baseline channel to the average speed of the simulated person in the initial exploration section is used as the speed reduction factor. The chaotic drag of the bypass path is obtained by linearly weighting and summing the directional sway factor and the velocity reduction factor. The traffic resistance coefficient is obtained by adding the geometric resistance of the detour path to the chaotic resistance of the detour path.

8. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 6, characterized in that, The process of determining the traffic resistance coefficient based on the geometric resistance and occupancy data of the detour route is as follows: Calculate the ratio of the intrusion width of the decoration component to the original width of the reference passage, and use it as the width occupancy factor; Calculate the ratio of the intrusion length of the decoration component to the actual length of the detour path, and use it as the length occupancy factor; Multiplying the width occupancy factor by the length occupancy factor yields the resistance affected by channel occupancy. The traffic resistance coefficient is obtained by adding the geometric resistance of the detour path to the resistance caused by the occupancy of the passage.

9. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 1, characterized in that, The optimization of the construction parameters, with the goal of minimizing the traffic resistance coefficient, involves the following process for optimizing the parameter set of the planned decoration components: The baseline evacuation route section between the abnormal node and the regression point is sliced ​​at a set interval to obtain several slices perpendicular to the center line of the baseline evacuation route. The parameter optimization condition is that the sum of the width of the adjustable occupancy range, the depth of the adjustable occupancy range, the installation center coordinate range, and the width of all planned decoration components in the same slice is not greater than the difference between the reference net width of the passage and the minimum net width of evacuation in the fire protection code. Determine the priority of each planned decoration component; For the first planned decoration component to be optimized, within the adjustable width range, the step size is adjusted according to preset parameters to generate alternative width values, forming a set of feasible width domains. Traverse all candidate values ​​for occupied width within the feasible region set for occupied width, and simultaneously generate candidate values ​​for occupied depth by adjusting the step size according to preset parameters within the adjustable range for occupied depth, thus forming a two-dimensional feasible region set for parameters; Traverse all parameter combinations within the feasible domain of two-dimensional parameters, and simultaneously generate alternative values ​​for the installation center coordinates by adjusting the step size according to the preset parameters within the adjustable range of the installation center coordinates, thus forming a three-dimensional parameter combination. For each combination of three-dimensional parameters, calculate the corresponding traffic resistance coefficient; The combination of three-dimensional parameters corresponding to the minimum traffic resistance coefficient is used as the optimized parameters of the current planned decoration component to be optimized; Continue to optimize the next priority planning and decoration components.

10. The intelligent evaluation method for public space decoration layout based on safe evacuation as described in claim 9, characterized in that, The process of determining the priority of each planned decoration component is as follows: If there are multiple planned decoration components on a certain slice, the width occupancy factor of each planned decoration component at the slice is calculated separately. The width occupancy factor represents the degree of occupancy impact of a single channel, and the priority is determined according to the degree of occupancy impact of a single channel from large to small. If there are multiple planned decoration components, but no multiple planned decoration components exist on any slice, then the priority is determined according to the distance between the planned decoration component and the starting point of the evacuation route, from largest to smallest, with the larger the distance, the higher the priority. If there is only one planned decoration component, then it is directly identified as the object to be optimized.

Images