Architectural interior design layout optimization method

Abstract

The invention provides a building interior design layout optimization method, and relates to the technical field of indoor modeling and layout optimization, and the method comprises the steps: obtaining building structure boundary data and current situation layout information, constructing an indoor three-dimensional space model, and completing the marking of node passing attributes and a spatial topological structure. On the basis, crowd behavior data of a daily scene and an emergency scene are collected, behavior characteristics such as walking speed, obstacle avoidance response and congestion sensitivity are extracted, and a walking characteristic model used for simulating the individual motion trend is established. And performing multi-scene simulation under a daily event and an emergency evacuation event by using the walking feature model, calculating performance indexes such as passing time, local congestion degree, key node congestion probability and total evacuation time, and identifying a congestion sensitive area. And further constructing a multi-target evaluation model including traffic efficiency, congestion degree and evacuation performance, and performing global optimization on layout adjustable parameters by adopting an algorithm to obtain a candidate layout scheme set.

Description

Technical Field

[0001] This invention relates to the field of interior modeling and layout optimization technology, and in particular to a method for optimizing the layout of architectural interior design. Background Technology

[0002] With the increasing number of large public buildings, commercial complexes, and multifunctional office spaces, interior space structures are becoming increasingly complex and functional requirements are highly diverse. Traditional architectural interior design methods are no longer sufficient to meet the safety and efficiency requirements of high-density environments. Existing technologies typically rely on two-dimensional floor plans or static rules to empirically configure passageway widths, furniture arrangements, and safe evacuation routes, but this method has the following shortcomings: Firstly, indoor spaces exhibit significant dynamism during actual use. The walking speed, obstacle avoidance behavior, and congestion response of people in different scenarios will exhibit complex dynamic characteristics as the spatial structure and population density change. Static rules cannot characterize these behavioral differences, resulting in a serious disconnect between design results and actual usage. Secondly, most existing interior design tools only support simulation analysis of single scenarios (such as normal passage or single-scenario evacuation), making it difficult to handle multiple event scenarios such as peak commuting, meeting dispersal, partial closure, and emergency evacuation at the same time. They are unable to systematically identify key bottlenecks and congestion-sensitive areas in each scenario, making it difficult for design schemes to balance daily efficiency and safety requirements in extreme situations. Therefore, we propose a method for optimizing the layout of building interior design.

[0003] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for optimizing the layout of building interior design, thereby solving the technical problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for optimizing the layout of building interior design includes the following steps: S1. Obtain the structural boundary data, functional area division data, and existing furniture and partition layout information of the target building. Construct an indoor three-dimensional space model based on a unified indoor space coordinate system, and generate a spatial mesh containing passages, obstacles, and arrangable areas in the indoor three-dimensional space model for subsequent crowd flow simulation and layout feasibility determination. S2. In the indoor three-dimensional spatial model, collect typical crowd behavior data of the target building. Crowd behavior data includes daily walking trajectories, time spent in the area, and movement preferences in emergency situations. Based on the crowd behavior data, construct a walking feature model to describe crowd movement trends, yielding behavior, and congestion response behavior. S3. Based on the walking feature model, generate a multi-event simulation set including daily use scenarios and emergency evacuation scenarios in the indoor three-dimensional space model; in each simulation event, calculate the passage path, passage time, local congestion degree, evacuation time and key node congestion probability of different individuals according to the current arrangement of furniture, partitions and passages, so as to obtain the crowd flow performance index under the layout. S4. In the indoor three-dimensional space model, set the furniture position, partition position, passage width, opening and closing status of passage nodes and the movable range of obstacles as layout adjustable parameters; based on the crowd mobility performance index, construct a multi-objective evaluation model with the goals of "maximizing daily walking efficiency, minimizing regional congestion, and minimizing emergency evacuation time" and the constraints of "functional zoning boundaries cannot be crossed and building structure cannot be moved". S5. Use reinforcement learning algorithm, genetic algorithm or other intelligent search algorithm to perform global search on the adjustable parameters of the layout; during the search process, dynamically adjust the parameter value range of the next search based on the multi-objective evaluation results obtained in each round of simulation, so that the layout gradually converges towards the direction of meeting the optimal combination of traffic efficiency, congestion intensity and evacuation performance, and obtain a set of candidate layout schemes that meet the requirements of multiple scenarios. S6. Based on the preset safety weight coefficient and functional weight coefficient, give a comprehensive score to each layout scheme in the candidate layout scheme set; select the layout with the highest comprehensive score as the final design layout of the target building, and output furniture arrangement scheme, partition arrangement scheme and key passage optimization suggestions.

[0006] S1 specifically involves: acquiring structural boundary data of the target building, including load-bearing walls, non-load-bearing walls, doors and windows, beams and columns, and functional area boundaries, and uniformly converting the structural boundary data to the indoor space coordinate system; collecting existing furniture positions, partition positions, and passageway width layout information, and fusing the layout information with the structural boundary data to form an initial layout dataset for the indoor space; based on the initial layout dataset, constructing an indoor 3D space model containing passageway nodes, obstacle nodes, and deployable nodes, with nodes established based on a spatial discrete grid for subsequent crowd flow simulation; labeling each spatial node in the indoor 3D space model with accessibility attributes, including accessible nodes, restricted access nodes, and inaccessible nodes, to ensure that the path planning process for crowd flow simulation has clear spatial constraints; based on accessibility attributes, generating adjacency relationships between each accessible node, constructing a simulation basic topology structure to describe spatial accessibility, and providing a basis for path search in subsequent multi-event simulations.

[0007] S2 specifically involves: acquiring crowd behavior data of the target building under different usage scenarios, including daily walking trajectories, dwell time in areas, and turning preferences, and generating a raw behavior dataset; collecting motion behavior data under emergency conditions through drills or simulations, including acceleration response, obstacle avoidance reaction time, and stagnation characteristics in congested areas, forming an emergency behavior dataset; cleaning, extracting features, and performing time-series modeling on the raw behavior dataset and the emergency behavior dataset to obtain a set of behavioral features representing individual movement trends, yielding behavior, and congestion response; inputting the set of behavioral features into a preset model building algorithm (rule-based modeling or machine learning-based modeling) to establish a walking feature model for characterizing crowd movement trends and behavioral decisions; applying the walking feature model to known crowd movement scenarios, comparing the consistency between simulated trajectories and real trajectories to verify the effectiveness of the model parameters, and correcting the model parameters when the deviation exceeds a threshold.

[0008] Specifically, S3 involves: constructing a multi-event simulation set in an indoor 3D space model based on a walking feature model, encompassing both daily usage events and emergency evacuation events. This event set covers different usage conditions of the building. In each simulation event, the walking feature model is used to calculate the movement path and path selection probability of each individual based on the current furniture position, partition position, and passage width parameters. Based on the movement path, the system calculates travel time, local congestion level, congestion probability at key nodes, and crowd mobility performance indicators for emergency evacuation, quantifying the traffic efficiency and safety performance of the current layout. Based on the crowd mobility performance indicators, areas where the congestion probability exceeds a threshold in different simulation events are identified and marked as congestion-sensitive areas to guide subsequent layout optimization. The consistency of the simulation results of the multi-event simulation set under different layout parameters is checked to ensure the stability and reliability of the crowd mobility performance indicators output by the simulation.

[0009] Specifically, S4 involves defining furniture positions, partition positions, passageway widths, node opening and closing states, and obstacle movement ranges as adjustable layout parameters in the indoor 3D spatial model to describe the layout adjustment space. Based on building functional zoning boundaries, structurally immovable areas, and fire safety regulations, insurmountable and immovable constraints are constructed to ensure the layout optimization process meets usage and safety requirements. A multi-objective evaluation index system is built, encompassing "daily walking efficiency, local congestion, emergency evacuation time, and probability of congestion at key nodes," based on pedestrian mobility performance indicators. A multi-objective evaluation model is established, centered on "maximizing traffic efficiency, minimizing congestion intensity, and optimizing evacuation performance," based on the evaluation index system and constraints. The effectiveness of the objective function and constraints is verified by running the multi-objective evaluation model on several typical layout samples, and the model weight parameters are adjusted when the evaluation results do not meet expectations.

[0010] S5 specifically involves: within the range of adjustable layout parameters, randomly initializing the layout search population or search points to form an initial solution set for layout optimization; inputting the initial solution set into the multi-event simulation set one by one to obtain the crowd flow performance index corresponding to each layout, and calculating its comprehensive evaluation value under the multi-objective evaluation model; based on the comprehensive evaluation value, generating the next round of adjustable layout parameter value set through reinforcement learning algorithm, genetic algorithm, or other intelligent search algorithm to make the layout search process approach a better solution region; determining whether the change amplitude of the comprehensive evaluation value in consecutive rounds of search is lower than the preset threshold; if it is lower than the threshold, the search process is determined to have converged; otherwise, the next round of search continues; after the search process converges, summarizing the layout schemes that meet the comprehensive optimization requirements of traffic efficiency, congestion intensity, and evacuation performance to form a candidate layout scheme set.

[0011] S6 specifically involves: setting traffic efficiency weighting coefficients and evacuation safety weighting coefficients based on building function, usage requirements, and safety level to comprehensively evaluate candidate layout schemes; using weighting coefficients and a multi-objective evaluation model to calculate a comprehensive score for each scheme in the candidate layout scheme set; selecting the layout scheme with the highest comprehensive score as the optimal layout scheme for the target building, and determining the furniture placement, partition placement, and passageway optimization suggestions corresponding to the layout scheme; mapping the optimal layout scheme to an interior 3D space model and generating 2D floor plans and 3D display diagrams to demonstrate the layout optimization effect; and outputting furniture adjustment suggestions, partition optimization suggestions, and key passageway handling strategies corresponding to the optimal layout scheme to guide subsequent construction and space renovation.

[0012] The beneficial effects of this invention are as follows: This invention constructs a 3D indoor spatial model by discretizing nodes and annotating access attributes and topological connections. This enables precise digital representation of internal building passageways, furniture, partitions, and obstacles, providing a structured and computable spatial foundation for subsequent simulations. Compared to traditional static designs relying on 2D drawings, this method offers higher accuracy and reproducibility. By collecting crowd behavior data under daily and emergency conditions and constructing a walking characteristic model, it can simulate dynamic behavioral characteristics such as expected speed, obstacle avoidance responses, and congestion stress. This upgrades indoor layout assessment from static rule review to simulation calculations driven by real-world behavior, significantly improving the matching degree of layout assessment to real-world usage scenarios.

[0013] This invention performs simulations under various events, including peak hours, meeting dismissal, partial lockdowns, and emergency evacuations, to calculate core performance indicators such as travel time, congestion level, total evacuation time, and congestion probability in key areas. It can also automatically identify congestion-sensitive areas, achieving multi-scenario layout performance evaluation capabilities that traditional design methods cannot reach. A comprehensive evaluation function integrating traffic efficiency, congestion indicators, evacuation performance, and key node risks is constructed. An adjustable weight mechanism enables adaptive optimization strategies for different building functions, allowing layout decisions to no longer rely on experience but rather on interpretable and quantifiable comprehensive performance scores.

[0014] This invention utilizes intelligent optimization strategies such as genetic algorithms, simulated annealing, and reinforcement learning to perform global searches and multiple iterations on a large number of layout schemes, automatically approximating the optimal solution. This avoids common problems in traditional layout adjustments, such as "local optima" and "repeated manual trial and error," significantly improving the efficiency and accuracy of layout optimization. The final output is the optimal layout scheme, generating floor plans, 3D renderings, pedestrian flow heatmaps, and evacuation route maps. It also provides detailed suggestions for furniture movement, partition adjustments, and passageway optimization, achieving a closed-loop process from simulation analysis to actual construction, offering directly actionable optimization guidance for designers and construction companies. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a method for optimizing the layout of building interior design according to the present invention. Detailed Implementation

[0016] 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.

[0017] Example 1: As Figure 1As shown in the figure, this embodiment provides a method for optimizing the layout of building interior design, including the following steps: S1. Indoor space modeling steps: Obtain the structural boundary data, functional area division data, and existing furniture and partition layout information of the target building. Construct an indoor three-dimensional space model based on a unified indoor space coordinate system, and generate a spatial mesh containing passages, obstacles, and arrangeable areas in the indoor three-dimensional space model for subsequent crowd flow simulation and layout feasibility determination. S2. Crowd Behavior Collection and Walking Feature Modeling Steps: In the indoor 3D spatial model, collect typical crowd behavior data of the target building. Crowd behavior data includes daily walking trajectories, time spent in the area, and movement preferences in emergency situations. Based on the crowd behavior data, construct a walking feature model to describe crowd movement trends, yielding behavior, and congestion response behavior. S3. Scene event-driven crowd flow simulation steps: Based on the walking feature model, generate a multi-event simulation set including daily use scenarios and emergency evacuation scenarios in the indoor three-dimensional space model; In each simulation event, calculate the passage path, passage time, local congestion degree, evacuation time and key node congestion probability of different individuals according to the current arrangement of furniture, partitions and passages, so as to obtain the crowd flow performance index under the layout. S4. Adjustable Layout Parameter Setting and Multi-Objective Evaluation Steps: In the indoor 3D space model, the furniture position, partition position, passage width, opening and closing status of passage nodes, and movable range of obstacles are set as adjustable layout parameters; based on the crowd mobility performance index, a multi-objective evaluation model is constructed with the objectives of "maximizing daily walking efficiency, minimizing regional congestion, and minimizing emergency evacuation time" and the constraints of "functional zoning boundaries cannot be crossed and building structures cannot be moved". S5. Layout global optimization steps based on simulation feedback: Use reinforcement learning algorithm, genetic algorithm or other intelligent search algorithm to perform global search on the adjustable parameters of the layout; during the search process, dynamically adjust the parameter value range of the next search based on the multi-objective evaluation results obtained in each round of simulation, so that the layout gradually converges towards the direction of meeting the comprehensive optimal requirements of traffic efficiency, congestion intensity and evacuation performance, and obtain a set of candidate layout schemes that meet the requirements of multiple scenarios. S6. Optimal Layout Determination and Output Steps: Based on the preset safety weight coefficient and functional weight coefficient, each layout scheme in the candidate layout scheme set is comprehensively scored; the layout with the highest comprehensive score is selected as the final design layout of the target building, and furniture arrangement scheme, partition arrangement scheme and key passage optimization suggestions are output.

[0018] S1 specifically includes the following sub-steps: S110, Structural Boundary Data Acquisition Sub-step: Acquire the structural boundary data of the target building, including load-bearing walls, non-load-bearing walls, columns, beams, doors and windows, and functional area boundaries.

[0019] The structural boundary data is obtained through any of the following methods: BIM model, laser SLAM scan, 3D laser point cloud data, or CAD architectural drawings.

[0020] The above structural boundary data were uniformly converted to the indoor spatial coordinate system. ,in, It is an indoor spatial coordinate system (representing the positional reference of three-dimensional coordinates within the building space), with the origin of the coordinate system set at the building entrance or determined by the BIM model reference point.

[0021] S120, Current Layout Data Integration Sub-step: Collect information on existing furniture locations, partition locations, passageway widths, and obstacle distributions to form a layout data set. Then, fuse this layout data set with structural boundary data to generate an initial layout dataset. ,in, This is the initial layout dataset (a set of data representing the current layout of furniture, partitions, and obstacles in the building).

[0022] The data fusion process uses a unified data structure, which can be a parameterized JSON table, a BIM parameter set, or a list of scene objects.

[0023] S130, Indoor 3D Spatial Discretization Sub-step: Based on the initial layout dataset Construct an indoor 3D space model and follow the preset mesh size. Discretize the interior space to generate a set of nodes. ,in, Representing spatial discrete grids in Length step in the direction; Representing spatial discrete grids in Length step in the direction; Representing spatial discrete grids in The height step size in the direction; (the values ​​of the three parameters range from 0.1m to 1.0m). It represents a set of spatial nodes, each node containing spatial coordinates, a node type label, and the node's reachability status.

[0024] Node type labels are used to distinguish categories such as "empty node, obstacle node, and boundary node".

[0025] S140, Passage attribute annotation sub-step: For each spatial node Mark the access attribute value ,in, Indicates the first One spatial node; This represents the node's accessibility attribute value.

[0026] The rules for access attribute values ​​are as follows: =1: Passable node; =0.5: Conditions for passage (e.g., nodes where openable or closable barriers or one-way passage facilities are located); =0: Impassable nodes (such as nodes containing load-bearing walls or fixed obstacles).

[0027] After annotation, a node attribute matrix is ​​generated. ,in, This is a node attribute matrix (representing the set of all node access states).

[0028] S150, Spatial Topology Generation Sub-step: Based on the node attribute matrix Construct a spatial topology graph ,in, A spatial topology graph (used to describe the reachability relationships between nodes); For a set of nodes (i.e.) ); Let be the set of edges (representing the walkable connections between any two nodes).

[0029] The rules for generating adjacency relationships are as follows: If any two nodes and Satisfy its Euclidean distance: Then at the node and Establish a topological edge between them And add the topological edge to the edge set. .in, Indicates that by node With nodes The topological edges formed; This represents the Euclidean distance between nodes.

[0030] After construction, the spatial topology map Used for path planning and regional accessibility calculation in subsequent crowd flow simulations.

[0031] S2 specifically includes the following sub-steps: S210, Daily Behavior Data Collection Sub-step: Inside the target building, collect daily behavior data of personnel through camera tracking, WiFi movement trajectory recording, infrared people detection equipment, or Bluetooth positioning equipment to form a raw daily behavior dataset. ,in, This is a daily behavior dataset (representing a collection of behavioral data such as movement trajectories, speed changes, and dwell distribution of people in normal usage scenarios).

[0032] The data collected includes: daily walking trajectories, typical turning behaviors, distribution of stationary positions, actual frequency of passageway use, and changes in the direction of pedestrian flow at different times.

[0033] S220, Emergency Behavioral Data Collection Sub-step: Collect behavioral data of personnel in an emergency situation through evacuation drills, simulator recording, or emergency broadcast triggering experiments to form an emergency behavior dataset. ,in, This is an emergency behavior dataset (a collection of data representing people's acceleration response, obstacle avoidance reaction, congestion and stagnation behavior, etc. in emergency situations).

[0034] The collected behaviors include: acceleration change sequence, obstacle avoidance reaction time, evacuation direction deviation characteristics, local stagnation behavior when congestion occurs, and speed decrease characteristics near danger zones.

[0035] S230, Behavioral Feature Extraction Sub-step: Extracting daily behavioral datasets With emergency behavior dataset Filtering, noise suppression, trajectory segmentation and temporal alignment are performed, and behavioral feature sets are extracted. ,in, The behavioral feature set (representing the set of feature variables used to describe pedestrian movement trends and reaction behaviors); For the first Individual behavioral characteristics (representing characteristic values ​​such as rate of change of velocity, angular velocity, and reaction time).

[0036] The specific features extracted include: walking speed change rate (acceleration and deceleration characteristics), turning angular velocity, obstacle avoidance reaction time, congestion sensitivity, probability of standing still, and area preference coefficient.

[0037] All eigenvalues ​​are standardized to ensure consistency in the dimensions of different eigenvalues.

[0038] S240, Walking Feature Model Construction Sub-step: Based on Behavioral Feature Set Constructing a walking feature model ,in, The walking characteristic model (represents a mathematical model used to simulate the movement trends, obstacle avoidance behavior, and congestion response behavior of an individual).

[0039] Walking feature model It is described using dynamic equations, specifically defined as follows: In the formula: For individual quality; The vector of the actual velocity of the individual at time t; The desired speed (corresponding to the desired speed term described in S240); This is the unit direction vector pointing towards the target point; The response relaxation time (representing the time required for an individual to adjust their speed to the desired speed); The obstacle avoidance force term (represents the vector sum of repulsive forces generated by an individual to avoid obstacles or other people in front of them, usually with a negative exponential relationship with distance); The congestion pressure term represents the vector sum of physical contact forces and psychological frictional forces experienced by an individual in a high-density area due to compression and obstruction. These three terms, together through this dynamic equation, constitute a comprehensive description of the individual's movement trend, used for subsequent simulations.

[0040] S250, Model Parameter Calibration Sub-step: Adjust the walking feature model... The model is applied to at least one known scenario for comparative simulation. By comparing the consistency between the real trajectory and the simulated trajectory, the model parameters are calibrated.

[0041] Define trajectory error for: in Trajectory error (representing the difference between the simulated trajectory and the actual trajectory); This is a sequence of actual trajectories (representing the actual collected walking trajectory data of the crowd); For the simulated trajectory sequence (representing the walking feature model) (Generated simulated trajectory data).

[0042] If the error satisfies: Then the parameter calibration is considered qualified, among which This is the error threshold (representing the maximum allowable calibration error).

[0043] If the requirements are not met, the model parameters will be automatically adjusted and the simulation will be repeated until the error requirements are met.

[0044] S3 specifically includes the following sub-steps: S310, Sub-step for constructing a multi-event simulation set: Based on the walking feature model (Walking Feature Model) and Spatial Topology Map (Spatial topology diagram), constructing a multi-event simulation set ,in, The set of simulation events (representing the set of events used to simulate different usage states within a building); For the first Each simulation event (representing a simulated crowd movement event in a specific scenario).

[0045] Multi-event simulation set Includes the following two types of events: Daily event collection Used to simulate crowd behavior in non-emergency situations, including: rush hour events, lunchtime movement events, and meeting break events.

[0046] Emergency Collection This is used to simulate crowd behavior during emergency situations, including fire evacuation events, alarm evacuation events, and partial lockdown detour events. Therefore, a set of multiple events can be represented as: S320, Sub-step for calculating individual path and migration probability: In any simulation event In this process, a path and motion sequence are generated for each individual participating in the simulation.

[0047] Assume there is a common set of individuals. ,in, For the set of individuals (representing all pedestrians participating in the simulation); For the first Individual.

[0048] Based on spatial topology graph Walking feature model Using Dijkstra's algorithm or Heuristic search algorithms for each individual calculate: Optimal path set : For individuals The m-th candidate walking path (consisting of a sequence of nodes).

[0049] Path selection probability Define path The geometric length is The probability that an individual chooses the m-th path is: in For individuals The probability of choosing its m-th path; This represents the length of the m-th path. For individuals The set of indices for all candidate paths; This is the distance sensitivity coefficient (representing how sensitive pedestrians are to path length). Let be the length of the j-th candidate path.

[0050] Real-time motion sequences : For individuals In time Spatial location sequence at time; individual movement sequence composed of walking feature model Control (the combined effects of desired speed, obstacle avoidance force, and congestion pressure).

[0051] S330, Sub-step for calculating crowd mobility performance index: In the event In, based on the motion sequences of all individuals Calculate the following performance metrics: (1) Travel time indicators : in, Travel time (representing the time required for an individual to complete the path); Start time; This is the end time.

[0052] (2) Local congestion index Let the spatial node be... ,but: in, For nodes At any moment Local congestion level; For nodes At any moment The number of people; For nodes The capacity (representing the maximum number of people that the node can accommodate).

[0053] (3) Congestion probability at key nodes : in, This represents the probability of congestion at critical nodes (indicating the probability that the congestion level exceeds a threshold). The congestion threshold (e.g., 0.8).

[0054] (4) Total time for emergency evacuation : in, Total emergency evacuation time (representing the time it takes for the last person to reach a safe exit); For individuals The passage time.

[0055] S340, Congestion-Sensitive Area Identification Sub-Step: Statistical Analysis of Local Congestion Degree Sequence The recognition satisfies: All nodes The set of regions formed by these nodes is defined as congestion-sensitive areas: in, This is a set of congestion-sensitive areas (representing spatial areas prone to congestion).

[0056] S350, Simulation Performance Stability Verification Sub-step: In multiple simulation events The stability of the performance indicators will be verified.

[0057] Define performance variability: in, This refers to the fluctuation in travel time. This represents the fluctuation in congestion levels. , The maximum / minimum passage time for each event; , This represents the maximum / minimum congestion level for each event.

[0058] If the following conditions are met: , The simulation performance is then considered stable. This is the threshold for fluctuations in travel time; This is the threshold for congestion fluctuation.

[0059] S360. Output declaration for this step: Output the following simulation result item: Travel time. Local congestion sequence Congestion probability at key nodes Total emergency evacuation time Congestion-sensitive areas This provides a quantitative basis for subsequent layout optimization.

[0060] S4 specifically includes the following sub-steps: S410, Layout Adjustable Parameter Definition Sub-step: Based on the indoor 3D spatial model and candidate layout areas, define the layout adjustable parameter vector. ,in, This is a vector of adjustable layout parameters (representing the set of all adjustable layout variables); For the first Adjustable layout parameters (such as furniture translation amount, partition movement amount, aisle width adjustment amount, etc.).

[0061] The specific categories of adjustable layout parameters include: furniture position adjustment parameters (such as translation distance and rotation angle); partition position and opening / closing status adjustment parameters; passage width adjustment parameters; and local obstacle movement range parameters.

[0062] The range of values ​​for each parameter is constrained by building codes, room boundaries, and structural limitations.

[0063] S420, Layout Constraint Construction Sub-step: To ensure the feasibility of the layout adjustment process, define a set of constraints. ,in, The set of constraints (representing the set of all restrictions that affect layout adjustments); For the first Constraints (such as structural or regulatory restrictions).

[0064] The constraints include: Structural immutability constraints: The positions of load-bearing walls, fixed columns, and main beams cannot be changed.

[0065] Minimum channel width constraint: in, This refers to the channel width; Minimum allowable channel width (e.g., 0.8m).

[0066] Functional boundaries must not be crossed: furniture and partitions must not be moved outside their respective functional areas.

[0067] Furniture movement range constraints: in, The distance that furniture can be moved (from...) ); The maximum allowable movement distance (e.g., 1.5m).

[0068] S430, Sub-step for constructing the evaluation index system: Based on the multi-event simulation output results, construct a multi-objective evaluation index system. ,in, This is a set of evaluation metrics (representing all metrics used to measure the quality of the current layout). For the first One evaluation indicator.

[0069] The evaluation index system should include at least four types of indicators: Daily traffic efficiency indicators : Daily traffic efficiency (representing the average traffic efficiency or average speed of all individuals in a daily event); local congestion index Defined as the average of all local congestion levels: in, For the first Local congestion at each node; This represents the number of nodes participating in the statistics.

[0070] Key node congestion probability index : This represents the congestion probability at key nodes (indicating the likelihood of congestion occurring at important nodes).

[0071] Emergency evacuation performance indicators : This represents the total emergency evacuation time (the time required for the last person to reach a safe exit). These metrics are used to evaluate the layout's daily usability and safety in emergency situations.

[0072] S440, Sub-steps for establishing a multi-objective evaluation model: Based on the set of evaluation indicators Construct a comprehensive evaluation function This is used to measure the quality of the current layout. The comprehensive evaluation function is defined as: in This is the overall evaluation value (representing the overall performance score of the layout); This is a weighting coefficient for daily traffic efficiency; This is the weighting coefficient for local congestion. This is the weighting coefficient for emergency evacuation performance; The congestion probability weighting coefficient for key nodes. A smoothing constant (e.g., 0.001) is used to prevent the denominator from being zero. This is a daily traffic efficiency indicator.

[0073] Weight parameters , , , It is used to achieve adaptive adjustment based on the building's usage scenario to prioritize either "efficiency" or "safety".

[0074] S450, Evaluation Model Feasibility Verification Sub-step: Using several typical layout samples... (From the set of adjustable layout parameters) Run the comprehensive evaluation function on ) Calculate its corresponding comprehensive evaluation value. ,in For the first The overall evaluation value of each layout sample.

[0075] The comprehensive evaluation results under different samples are compared and verified. If anomalies occur, such as: A certain indicator does not meet the constraints. The overall evaluation fluctuates significantly outside a reasonable range; the weight settings do not match the actual scenario; therefore, the weight parameters should be adjusted. , , , Adjustments were made to ensure the evaluation model was stable and met user needs.

[0076] S460. Output declaration for this step: Output multi-objective evaluation function. and all its indicator values , , , This provides a quantitative reference for the next step of global optimization.

[0077] S5 specifically includes the following sub-steps: S510, Optimize the search initialization sub-step: In the layout adjustable parameter vector Within the range of values, an initial layout parameter set is generated through random initialization or based on heuristic rules. ,in, This is the initial set of layout parameters (representing the set of layout parameter vectors used to initiate optimization). For the first A set of layout parameter vector samples (i.e., a set of layout schemes to be evaluated).

[0078] Initial layout parameter set The generation needs to satisfy a set of constraints. This ensures that all initial samples are feasible layouts.

[0079] S520, Simulation-driven Layout Evaluation Sub-step: For the initial layout parameter set Each layout parameter vector in Perform the following procedure: layout parameter vector Input multi-event simulation set By running a crowd flow simulation, a set of performance metrics for this layout were obtained: in For layout The set of evaluation indicators; For daily traffic efficiency; This represents the average congestion level. Probability of congestion at key nodes; This represents the total time for emergency evacuation.

[0080] set of indicators Input comprehensive evaluation function Calculate the overall evaluation value of this layout. ,in, For layout The overall evaluation score.

[0081] The formula for calculating the comprehensive evaluation value is: in, , , , To evaluate the weighting coefficients (representing the importance of different indicators), all calculation results are stored in the evaluation record set R, where R is the evaluation record set (representing the set of evaluation values ​​and indicator records for all layout samples).

[0082] S530, Search Strategy Update Sub-step: To gradually bring the layout closer to the optimal solution, based on the comprehensive evaluation value of the layout samples... The layout parameter set is updated using an intelligent optimization algorithm.

[0083] Optional algorithms include: Genetic Algorithm (GA), reinforcement learning policy iteration, simulated annealing (SA), and particle swarm optimization (PSO). Specifically, if a genetic algorithm is used, the layout adjustable parameter vector will be... Mapping to chromosome gene sequences using real-number encoding or binary encoding; if a reinforcement learning algorithm is used, the layout parameter vector... Defined as environment state, the parameter adjustment amount Defined as the Action space.

[0084] This step defines the "search update function". This represents a function that generates the next set of layout parameters based on the previous set of layout parameters.

[0085] A new set of parameters is generated by searching the update function: in, For the first The set of layout parameters for the wheels; For the first A new set of layout parameters; This is a comprehensive evaluation result of all layout samples from the previous round.

[0086] Search update function Its functions are: to enhance the weight of high-scoring samples (such as the selection operation of GA); to randomly perturb the parameters (such as the mutation operation); and to guide the parameters to search in the direction of improving the evaluation value (such as gradient boosting or policy iteration).

[0087] S540, Convergence Judgment Sub-step: After each round of search, determine whether the optimization has converged based on the change in the comprehensive evaluation value.

[0088] Define the maximum evaluation value sequence as ,in, For the first The highest overall evaluation value in the wheel layout set.

[0089] Calculate the evaluation gain over multiple consecutive rounds: in, For the first The evaluation gain of the round (representing the degree of change in the optimal value between two consecutive rounds); if the following conditions are met: Then it is considered convergent, where To evaluate the gain threshold (representing the minimum allowable gain range).

[0090] If the conditions are not met, continue to the next round of search.

[0091] S550, Candidate Layout Scheme Set Generation Sub-step: After the convergence condition is met, generate the final round of layout parameter set. Select all comprehensive evaluation values ​​that exceed the threshold. The layout forms a set of candidate layout schemes. ,in This is the layout evaluation threshold (representing the minimum performance standard that the layout must achieve). This is the set of candidate layout schemes (representing the set of layout parameters that meet the optimization requirements).

[0092] S560. Output declaration for this step: Output the following content: Final candidate layout scheme set The overall evaluation value of each candidate layout Key performance metrics set for each candidate layout R is the iterative record of the optimization process.

[0093] All outputs will be used in the final layout determination and output steps.

[0094] S6 specifically includes the following sub-steps: S610, Weighting Coefficient Setting Sub-step: Based on building usage requirements, functional characteristics, and safety levels, set the weighting coefficient vector for comprehensive evaluation. ,in, This is a weight coefficient vector (used to control the degree of influence of different evaluation indicators on the final score). This is a weighting coefficient for daily traffic efficiency; The average congestion index weighting coefficient; This is the weighting coefficient for emergency evacuation performance; The weighting coefficients for the congestion probability index at key nodes.

[0095] The weighting parameters are set according to strategies such as "efficiency first" or "safety first". For example: Office buildings can adopt Medical or high-security buildings may adopt .

[0096] S620, Sub-step for calculating comprehensive score: For the candidate layout scheme set Each layout parameter vector in Calculate its comprehensive evaluation value ,in, A set of candidate layout schemes; For the first A vector of candidate layout parameters; For the first The comprehensive evaluation value of each candidate layout scheme.

[0097] The formula for calculating the comprehensive evaluation value follows the multi-objective evaluation function defined in S520: in , , , All results are from S520 simulations.

[0098] All comprehensive evaluation values ​​are stored in the final evaluation set. ,in The final evaluation set (representing the comprehensive score set of all candidate layout schemes).

[0099] S630, Optimal Layout Selection Sub-step: In the final evaluation set Determine the layout scheme with the highest score. As the final optimal layout solution, among which... in, The optimal layout scheme (representing the layout parameter vector with the highest overall score).

[0100] During the selection process, the second and third best layouts can be output as backup options to support multiple comparative evaluations or subsequent manual review.

[0101] S640, Layout Visualization Sub-step: Presenting the optimal layout scheme Mapping to indoor 3D space model The system generates visualizations, including: two-dimensional floor plans (such as CAD floor plans); three-dimensional layout renderings (used to show the locations of furniture, partitions, and passageways); heat maps of key areas: generating congestion heat distribution maps using simulation data; evacuation route visualizations: showing evacuation routes and congestion hotspots in emergency situations; and visualizations used to assist in final construction, scheme review, and user decision-making.

[0102] S650, Optimization suggestion output sub-step: Based on the optimal layout scheme Based on the simulation data, output a final set of layout optimization suggestions. ,in, This is a set of layout optimization suggestions (a structured set of suggestions used to guide construction and final scene modification).

[0103] The recommendations include: Recommendations for furniture movement: such as the direction of movement, translation distance, and rotation angle for each piece of furniture.

[0104] Partition optimization suggestions: adjustments to partition movement, opening / closing status, and suggestions for opening / closing mechanisms.

[0105] Channel adjustment strategy: suggestions for widening the channel and rearranging bottlenecks.

[0106] Strategies for handling congestion-sensitive areas: setting up pedestrian detour routes; relocating temporary barriers; and proposing one-way traffic mechanisms.

[0107] Emergency evacuation optimization recommendations: key node identification; emergency route indicator light layout strategy; space release plan for congestion-prone areas.

[0108] The final output can be integrated with construction drawings, design drawings, and building management systems to achieve the most feasible intelligent layout adjustments.

[0109] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0110] 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, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0111] 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.

[0112] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0114] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0115] 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.

[0116] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0117] 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 scope of the technology 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.

[0118] In conclusion, 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 optimizing the layout of architectural interior design, characterized in that, Includes the following steps: S1. Obtain the structural boundary data, functional area division data, and existing furniture and partition layout information of the target building, construct an indoor three-dimensional space model, and generate a spatial mesh containing passages, obstacles, and arrangeable areas in the indoor three-dimensional space model for subsequent crowd flow simulation and layout feasibility determination. S2. In the indoor three-dimensional space model, collect typical crowd behavior data of the target building. Crowd behavior data includes daily walking trajectory, time spent in the area, and movement preferences in emergency situations. Based on crowd behavior data, a walking feature model is constructed to describe crowd movement trends, yielding behavior, and congestion response behavior. S3. Generate a multi-event simulation set including daily use scenarios and emergency evacuation scenarios in the indoor three-dimensional space model; calculate the passage path, passage time, local congestion degree, evacuation time and key node congestion probability of different individuals based on the current arrangement of furniture, partitions and passages, and obtain the crowd flow performance index under the layout. S4. Set the furniture position, partition position, passage width, opening and closing status of passage nodes and the movable range of obstacles as adjustable layout parameters; construct a multi-objective evaluation model with the goals of maximizing daily walking efficiency, minimizing regional congestion, and minimizing emergency evacuation time, and with the constraints of functional zoning boundaries not being crossed and building structures not being movable. S5. Use reinforcement learning algorithms, genetic algorithms or other intelligent search algorithms to perform a global search on the adjustable parameters of the layout; during the search process, dynamically adjust the parameter value range of the next search based on the multi-objective evaluation results obtained in each round of simulation, so that the layout gradually converges towards the direction of meeting the optimal combination of traffic efficiency, congestion intensity and evacuation performance, and obtain a set of candidate layout schemes that meet the requirements of multiple scenarios.

2. The method for optimizing the layout of building interior design according to claim 1, characterized in that, It also includes S6, which comprehensively scores each layout scheme in the candidate layout scheme set based on the preset security weight coefficient and the usage function weight coefficient; Select the layout with the highest overall score as the final design layout for the target building, and output furniture arrangement scheme, partition arrangement scheme and key passage optimization suggestions.

3. The method for optimizing the layout of building interior design according to claim 1, characterized in that, S1 specifically refers to: Obtain the structural boundary data of the target building and uniformly convert the structural boundary data to the indoor space coordinate system; Collect information on the location of existing furniture, partitions, and passageway widths, and integrate this information with structural boundary data to form an initial layout dataset for the interior space. Based on the initial layout dataset, an indoor three-dimensional spatial model is constructed, with nodes established based on a spatial discrete grid for subsequent crowd flow simulation. In the indoor 3D spatial model, each spatial node is labeled with access attributes to ensure that the path planning process of crowd flow simulation has clear spatial constraints; Based on accessibility attributes, adjacency relationships are generated between all accessible nodes to construct a basic simulation topology for describing spatial accessibility, providing a basis for path search in subsequent multi-event simulations.

4. The method for optimizing the layout of building interior design according to claim 1, characterized in that, S2 specifically refers to: Acquire crowd behavior data of the target building under different usage scenarios and generate raw behavior dataset; Collect motion behavior data under emergency conditions through drills or simulations to form an emergency behavior dataset. The original behavior dataset and the emergency behavior dataset are cleaned, feature extracted and time-series modeled to obtain a set of behavioral features that characterize individual movement trends, yielding behavior and congestion response. The behavioral feature set is input into a preset model building algorithm to establish a walking feature model for characterizing crowd movement trends and behavioral decisions. By applying the walking feature model to a known crowd movement scenario, the consistency between the simulated trajectory and the real trajectory is compared to verify the effectiveness of the model parameters, and the model parameters are corrected when the deviation exceeds the threshold.

5. The method for optimizing the layout of building interior design according to claim 1, characterized in that, S3 specifically refers to: Based on the walking feature model, a multi-event simulation set is constructed in the indoor three-dimensional space model. The event set is used to cover different usage conditions of the building. In each simulation event, the movement path and path selection probability of each individual are calculated using the walking feature model based on the current furniture position, partition position and aisle width parameters. Based on the movement path, the system calculates the travel time, local congestion level, congestion probability at key nodes, and crowd mobility performance indicators during emergency evacuation, which are used to quantify the traffic efficiency and safety performance of the current layout. Based on crowd mobility performance indicators, areas where the probability of congestion exceeds a threshold in different simulation events are identified and marked as congestion-sensitive areas to guide subsequent layout optimization. Consistency checks are performed on the simulation results of multi-event simulation sets under different layout parameters to ensure that the crowd flow performance indicators output by the simulation are stable and reliable.

6. The method for optimizing the layout of building interior design according to claim 1, characterized in that, S4 specifically refers to: In the indoor 3D space model, the furniture position, partition position, passage width, node opening and closing status and obstacle movement range are defined as layout adjustable parameters to describe the layout adjustment space; Based on the building's functional zoning boundaries, structurally immovable areas, and fire safety regulations, insurmountable and immovable constraints are constructed to ensure that the layout optimization process meets usage and safety requirements. Based on population mobility performance indicators, a multi-objective evaluation index system is constructed; Based on the evaluation index system and constraints, a multi-objective evaluation model is established with "maximizing traffic efficiency, minimizing congestion intensity, and optimizing evacuation performance" as its core. The effectiveness of the objective function and constraints of the multi-objective evaluation model was verified by running it on several typical layout samples, and the model weight parameters were adjusted when the evaluation results did not meet expectations.

7. The method for optimizing the layout of building interior design according to claim 1, characterized in that, S5 specifically refers to: Within the range of adjustable layout parameters, the layout search population or search points are randomly initialized to form the initial solution set for layout optimization. The initial solution set is input into the multi-event simulation set one by one to obtain the crowd flow performance index corresponding to each layout, and calculate its comprehensive evaluation value under the multi-objective evaluation model. Based on the comprehensive evaluation value, the set of values ​​for the next round of layout adjustable parameters is generated through reinforcement learning algorithms, genetic algorithms or other intelligent search algorithms, so that the layout search process approaches a better solution region; Determine whether the change in the comprehensive evaluation value in multiple consecutive search rounds is lower than the preset threshold. If it is lower than the threshold, the search process is considered to have converged; otherwise, continue to the next search round. After the search process converges, the layout schemes that meet the comprehensive optimization requirements of traffic efficiency, congestion intensity and evacuation performance are summarized to form a candidate layout scheme set.

8. The method for optimizing the layout of building interior design according to claim 2, characterized in that, S6 specifically refers to: Based on the building function, usage requirements and safety level, traffic efficiency weighting coefficient and evacuation safety weighting coefficient are set for comprehensive evaluation of candidate layout schemes. Using weighting coefficients and a multi-objective evaluation model, a comprehensive score is calculated for each scheme in the candidate layout scheme set; The layout scheme with the highest overall score is selected as the optimal layout scheme for the target building, and the furniture placement, partition placement, and passageway optimization suggestions corresponding to the layout scheme are determined.

9. The method for optimizing the layout of building interior design according to claim 8, characterized in that, S6 also includes: mapping the optimal layout scheme to the indoor 3D space model and generating 2D floor plan and 3D display diagram to show the layout optimization effect; outputting furniture adjustment suggestions, partition optimization suggestions and key passage handling strategies corresponding to the optimal layout scheme to guide subsequent construction and space renovation.

Images