Deeply buried metro entrance and exit elevator facility configuration method, product, equipment and medium

Abstract

The application provides a deep-buried subway entrance and exit elevator facility configuration method, product, equipment and medium. The method comprises the following steps: obtaining design input information of a target subway station, and constructing a pedestrian dynamic simulation model based on the design input information; the pedestrian dynamic simulation model comprises a fatigue model and a facility selection model; obtaining pre-set initial facility configuration parameters, and inputting the facility configuration parameters into the pedestrian dynamic simulation model for simulation running; based on the pedestrian space-time data collected in the simulation running process, a plurality of performance indexes corresponding to the initial facility configuration parameters are calculated; taking the plurality of performance indexes as optimization targets, taking the initial facility configuration parameters as decision variables, and adopting a multi-objective optimization algorithm to generate entrance and exit elevator facility configuration parameters of the target subway station. The configuration of the elevator and other facilities is changed from subjective judgment to an objective optimization process driven by simulation data, so that the scientificity, accuracy and efficiency of the deep-buried subway entrance and exit are improved.

Description

Technical Field

[0001] This application relates to the field of subway facility configuration technology, and more specifically, to a method, product, equipment and medium for configuring elevator facilities at subway entrances and exits. Background Technology

[0002] As urban subways expand deeper into underground stations, the vertical traffic pressure at station entrances and exits is significantly increasing. Current design methods primarily rely on static specifications and empirical formulas to determine the size and quantity of facilities such as stairs and escalators. There is a lack of clear quantitative guidelines regarding whether elevators should be installed in deeply buried conditions and how they should be configured. These methods largely depend on the subjective experience of designers and fail to consider the dynamic behavior of passengers during the design process. This results in potential discrepancies between existing design solutions and actual operational performance, hindering efficient design. Summary of the Invention

[0003] The purpose of this application is to provide a method, program product, electronic device and storage medium for configuring elevator facilities at deep-buried subway entrances and exits, so as to improve the above-mentioned problems.

[0004] In a first aspect, embodiments of this application provide a method for configuring elevator facilities at the entrances and exits of a deeply buried subway station, comprising: acquiring design input information of the target subway station, including predicted passenger flow data and station burial depth data; constructing a pedestrian dynamic simulation model based on the design input information; the pedestrian dynamic simulation model includes a fatigue model and a facility selection model; the fatigue model is used to simulate the speed decay of passengers walking on long staircases; the facility selection model is used to simulate the path selection of passengers among stairs, escalators, and elevators; acquiring pre-set initial facility configuration parameters and inputting the facility configuration parameters into the pedestrian dynamic simulation model for simulation operation; the initial facility configuration parameters include the specifications and quantity parameters of stairs, escalators, and elevators; calculating multiple performance indicators corresponding to the initial facility configuration parameters based on pedestrian spatiotemporal data collected during the simulation operation; the multiple performance indicators include passage time, queuing time, congestion density, and fatigue level; using the multiple performance indicators as optimization objectives and the initial facility configuration parameters as decision variables, and employing a multi-objective optimization algorithm, generating the elevator facility configuration parameters for the entrances and exits of the target subway station.

[0005] In the aforementioned implementation process, by inputting actual data such as predicted passenger flow and station burial depth into a pedestrian dynamic simulation model that integrates fatigue and facility selection models, it is possible to dynamically simulate and evaluate any given facility configuration parameter scheme, obtaining multiple performance indicators reflecting real-world operation, such as travel time, queuing time, congestion density, and fatigue level. Furthermore, using a multi-objective optimization algorithm, with these performance indicators as targets, the specifications and quantity parameters of stairs, escalators, and elevators are automatically searched and iteratively optimized on a large scale, ultimately generating a configuration scheme with superior overall performance. This method overcomes the limitations of traditional designs relying on static specifications and human experience, transforming the configuration of facilities such as elevators from subjective judgment to an objective optimization process driven by simulation data. This significantly improves the scientific rigor, accuracy, and efficiency of deep-buried subway entrance and exit design, helping to better balance construction costs with passenger efficiency, safety, and comfort while meeting traffic demands.

[0006] Optionally, in this embodiment of the application, the fatigue model is:

[0007]

[0008] in, For pedestrian speed, Let i be the expected speed of pedestrian i. The fatigue index. The distance traveled. For risk perception, For the perception of fatigue. This represents the minimum speed for pedestrians.

[0009] In the aforementioned implementation process, risk perception and fatigue perception are introduced as dynamic adjustment factors to simulate the speed decay of passengers walking on long staircases. The advantage of this approach is that it transforms the simulation of pedestrian fatigue from a fixed or time-dependent simple function into one that reflects changes in the passenger's psychological state in a real-world environment. For example, when sensing congestion or obstruction ahead, the risk perception factor may cause the passenger to proactively reduce speed; while after sustained physical exertion, the fatigue perception factor will dominate further speed decay. This dynamic, perception-driven speed calculation method makes the behavior of virtual passengers in the simulation more closely resemble the complex reactions of real humans walking on deeply buried staircases, thereby improving the realism and accuracy of the simulation model in simulating passenger physical exertion and speed changes.

[0010] Optionally, in this embodiment of the application, the facility selection model determines the probability of a pedestrian choosing a facility by: calculating the generalized cost of a pedestrian choosing each facility, wherein the generalized cost is determined by local cost and global cost; the local cost includes a queuing factor determined by the number of people queuing in front of the facility, and a distance factor for the pedestrian to reach the facility entrance; the global cost is the estimated time required for the pedestrian to reach their destination through the facility; and based on the generalized cost, the probability of a pedestrian choosing each facility is calculated by a regression selection model.

[0011] In the above implementation process, a generalized cost is calculated by comprehensively considering local waiting conditions, arrival distance, and total travel time, and a regression model is used to predict the selection probability based on this cost. This approach makes the passenger's choice behavior among stairs, escalators, and elevators in the simulation no longer a simple rule judgment, but rather a more realistic decision-making process for passengers under the weighting of multiple factors, thus significantly improving the behavioral realism of the facility selection model and the credibility of the simulation results.

[0012] Optionally, in this embodiment, multiple performance indicators include passage time, queuing time, congestion density, and fatigue level. Based on pedestrian spatiotemporal data collected during simulation, multiple performance indicators corresponding to the initial facility configuration parameters are calculated, including: passage time based on pedestrian entry and exit timestamps; queuing time based on pedestrian waiting time at each facility entrance; congestion density based on instantaneous pedestrian density at preset key nodes; and fatigue level calculated after normalization based on pedestrian height data changed by walking.

[0013] In the aforementioned implementation process, key design evaluation dimensions such as traffic efficiency, queuing conditions, congestion risk, and fatigue levels are transformed into specific, repeatable, and quantifiable indicators based on clear data sources and calculation rules. For example, travel time is calculated using timestamp differences, and queuing time is calculated using waiting time, which is intuitive and resistant to interference. By designing corresponding data collection and calculation paths for different phenomena, the evaluation of the design scheme is comprehensive and objective, covering different aspects from macro-efficiency to micro-experience. This standardized, multi-dimensional quantitative evaluation system provides a unified and reliable scale and basis for fairly comparing the merits of different facility configuration schemes.

[0014] Optionally, in this embodiment, multiple performance indicators are used as optimization objectives, initial facility configuration parameters are used as decision variables, and a multi-objective optimization algorithm is employed to generate the entrance and exit elevator facility configuration parameters of the target subway station. This includes: obtaining an initial population containing decision variables, where each individual in the initial population represents a set of initial facility configuration parameters; for each individual in the initial population, a pedestrian dynamic simulation model is invoked to perform simulation and calculate the corresponding multiple performance indicators; based on the multiple performance indicators and facility construction costs, the initial population is non-dominatedly sorted, and a new population is generated based on the sorting results; the new population is used as the initial population for iterative execution until the termination condition is met, generating a Pareto optimal solution set; and the entrance and exit elevator facility configuration parameters of the target subway station are determined from the Pareto optimal solution set.

[0015] In the aforementioned implementation process, a large number of different initial facility configuration parameter combinations are first generated as an initial population. Then, a pedestrian dynamic simulation model is automatically invoked to simulate each scheme and calculate multiple target indicators such as travel time, queuing time, congestion density, fatigue level, and cost. Next, based on these indicators, non-dominated sorting is used to classify and screen the schemes, and an evolutionary operation is used to generate a new population of schemes with better performance. This process is iterated repeatedly, and finally, a series of Pareto optimal schemes that achieve the best balance between multiple performance and cost objectives are automatically found for selection. This process transforms the originally experience-based and tedious manual calculation and scheme comparison work into a systematic automatic search and optimization process driven by computer simulation and intelligent algorithms. This improves the design efficiency and scientific nature of elevator facility configuration at deep-buried subway entrances and exits, and helps to better ensure traffic efficiency and passenger comfort while controlling construction costs.

[0016] Optionally, in the embodiments of this application, the decision variables include: staircase width, escalator width, number of elevators, and rated elevator speed.

[0017] Optionally, in this embodiment of the application, the design input information also includes elevator access capacity. Based on the design input information, a pedestrian dynamic simulation model is constructed, including: determining the parameters of the fatigue model based on the station burial depth data; initializing the weight parameters in the facility selection model based on the predicted passenger flow data; and constructing the pedestrian dynamic simulation model according to the parameters of the fatigue model and the weight parameters in the facility selection model.

[0018] In the above implementation process, firstly, the fatigue model is configured with targeted parameters using station burial depth data, enabling the model to accurately reflect the speed decay characteristics of passengers at different station depths. Simultaneously, the weight parameters of the facility selection model are initialized using predicted passenger flow data, ensuring that passenger decisions in the model align with behavioral patterns under actual passenger flow pressure. Finally, these two pre-configured core behavioral modules are integrated into the basic simulation engine to construct a highly customized pedestrian dynamic simulation model. This method ensures that the final simulation model is no longer general or idealized, but closely matches the specific burial depth conditions and passenger flow predictions of the target subway station, thus providing a highly reliable behavioral basis for the simulation evaluation of subsequent facility configuration schemes. The resulting optimized design scheme better meets the efficiency, safety, and comfort requirements of specific stations in real-world operating environments, improving the accuracy and reliability of the design.

[0019] Secondly, this application also provides a device for configuring elevator facilities at the entrances and exits of a deeply buried subway station, comprising: an information acquisition module for acquiring design input information of the target subway station, including predicted passenger flow data and station burial depth data; a simulation model module for constructing a pedestrian dynamic simulation model based on the design input information; the pedestrian dynamic simulation model includes a fatigue model and a facility selection model; the fatigue model is used to simulate the speed decay of passengers walking on long staircases; the facility selection model is used to simulate the path selection of passengers among stairs, escalators, and elevators; a simulation operation module for acquiring pre-set initial facility configuration parameters and inputting the facility configuration parameters into the pedestrian dynamic simulation model for simulation operation; the initial facility configuration parameters include the specifications and quantity parameters of stairs, escalators, and elevators; a performance index module for calculating multiple performance indicators corresponding to the initial facility configuration parameters based on pedestrian spatiotemporal data collected during the simulation operation; the multiple performance indicators include passage time, queuing time, congestion density, and fatigue level; and an optimization module for generating the elevator facility configuration parameters at the entrances and exits of the target subway station using multiple performance indicators as optimization objectives, the initial facility configuration parameters as decision variables, and a multi-objective optimization algorithm.

[0020] Thirdly, embodiments of this application also provide a computer program product, including computer program instructions, which are executed by a processor to perform the method provided in the first aspect or any implementation thereof.

[0021] Fourthly, embodiments of this application also provide an electronic device, including: a processor and a memory, the memory storing computer program instructions, which are executed by the processor to perform the method provided in the first aspect or any implementation thereof.

[0022] Fifthly, embodiments of this application also provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, perform the method provided in the first aspect or any implementation thereof.

[0023] This application provides a method, program product, electronic equipment, and storage medium for configuring elevator facilities at deeply buried subway entrances. By inputting actual data such as predicted passenger flow and station burial depth into a pedestrian dynamic simulation model integrating fatigue and facility selection models, it can dynamically simulate and evaluate any given facility configuration parameter scheme, obtaining multiple performance indicators reflecting real-world operation, such as travel time, queuing time, congestion density, and fatigue level. Furthermore, using a multi-objective optimization algorithm, with these performance indicators as objectives, it automatically performs large-scale search and iterative optimization of the specifications and quantity parameters of stairs, escalators, and elevators, ultimately generating a configuration scheme with superior overall performance. This method overcomes the limitations of traditional designs relying on static specifications and manual experience, transforming the configuration of facilities such as elevators from subjective judgment to an objective optimization process driven by simulation data. This significantly improves the scientific rigor, accuracy, and efficiency of deeply buried subway entrance design, helping to better balance construction costs with passenger efficiency, safety, and comfort while meeting traffic demands. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a method for configuring elevator facilities at a deeply buried subway entrance / exit, provided as an embodiment of this application;

[0026] Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0027] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this application.

[0029] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0030] With rapid urbanization, subways, as a major mode of public transportation, are experiencing increasingly larger station sizes and more complex passenger flow organization, placing higher demands on the design of the number, location, and capacity of entrances and exits. To meet the needs of safe evacuation, passenger convenience, and coordinated development with urban space, existing technologies have proposed a subway entrance and exit design method based on multi-source data and 3D models, which is used to evaluate and construct station entrance and exit schemes in advance during the design phase.

[0031] In existing technologies, designers first need to determine the required number of subway station entrances and exits based on the surrounding urban environment, building planning, and passenger flow patterns. To this end, existing technologies typically use oblique photography by drones to acquire 3D real-world data of the station, and combine this with government-released information on surrounding building development plans to generate a high-precision 3D model reflecting the actual ground environment. Simultaneously, historical passenger flow data analysis and prediction models are used to obtain the station's potential future passenger volume, which serves as the basis for calculating the number of entrances and exits.

[0032] In determining the location of entrances and exits, existing technologies rely on borehole survey data to construct a three-dimensional geological model. By analyzing the underground soil structure, rock strata distribution, and groundwater level, suitable excavation areas are selected. Simultaneously, by combining predicted passenger flow, destinations, and traffic flow lines, the service radius on the ground is analyzed to determine multiple alternative entrance and exit locations. Based on this, existing technologies compare the feasibility of multiple locations to form the final entrance and exit layout plan.

[0033] After the location scheme is formed, the existing technology uses an improved Dijkstra algorithm to optimize the walking path between the entrance / exit and the turnstile. This algorithm treats the entrance / exit and the turnstile as nodes in a graph, and uses the path distance, congestion level, passage time and traffic convenience as attributes of the edges. By comprehensively scoring, it calculates the optimal path between each node, thereby obtaining the optimal passage route for passengers to walk within the station.

[0034] Based on the optimal path, existing technologies further require determining the width of entrances and exits and the configuration of stairs and escalators. This process calculates the flow capacity of each entrance and exit based on predicted passenger flow, design specifications, and evacuation time requirements, thereby determining the width of the entrances and exits, and configuring the number and type of stairs or escalators according to relevant standards to ensure that passenger evacuation needs are met.

[0035] Finally, existing technology constructs a three-dimensional structural model of the entrance / exit by sequentially arranging and splicing the models of escalator sections, passageways, and civil defense sections. By repeating this process, multiple subway entrance / exit models adapted to the station environment can be generated, providing a foundation for subsequent construction design and scheme demonstration.

[0036] The shortcomings of existing technologies include: (1) lack of design for elevator facilities under "deep burial conditions". As urban rail transit develops to deeper and more levels, more and more subway stations are showing "deep burial" characteristics due to urban terrain conditions and the trend of spatial longitudinal expansion. However, existing technical solutions and relevant national design specifications still lack clear guiding principles for the installation of elevator facilities under deep burial conditions. Whether to install elevators, the number of elevators and their key parameters mostly depend on the designer's experience judgment, lacking quantifiable design basis, making it difficult to form a unified standard for when to build elevators and how to configure elevators under different burial depths. The embodiments of this application introduce measurable and evaluable quantitative indicators through simulation technology, which can calculate and provide feedback on the necessity of elevator construction and configuration parameters under different burial depths, thereby providing a scientific decision-making basis for elevator design in deep burial stations.

[0037] (2) The configuration of entrance and exit facilities lacks a quantitative optimization mechanism and relies solely on the static verification of a single scheme. Existing technologies typically rely on standard provisions and empirical formulas to statically calculate the width and number of stairs and escalators in terms of entrance and exit facility configuration, and generate a single design scheme accordingly. This approach lacks the ability to compare multiple schemes, does not consider the parameter settings of vertical transportation facilities such as elevators, and cannot quantitatively evaluate key experience indicators such as "passage time, queuing time, crowd density, and fatigue level" under actual operating conditions. Therefore, existing technologies struggle to optimize the comprehensive performance of different facility combinations. The embodiments of this application, based on the introduction of simulation, can obtain dynamic experience indicators of passengers under specific facility configuration conditions, such as passage time, queuing time, density, and fatigue level; and utilize multi-objective optimization algorithms to optimize facility configuration parameters, thereby achieving optimized layout of entrance and exit facilities based on quantifiable indicators, making the final configuration scheme more scientific and reasonable.

[0038] Without the introduction of dynamic simulation methods, it is impossible to assess congestion, queuing, and traffic efficiency under real-world operational conditions. Existing technologies generally rely on static path optimization results as the primary basis for entrance and exit design, lacking the ability to dynamically simulate real-world operational conditions. Therefore, they can only obtain theoretical traffic capacity and cannot reflect real-world situations such as congestion, queuing, and traffic efficiency during peak hours, making it difficult to accurately assess the advantages and disadvantages of different schemes and the load distribution among facilities. Furthermore, existing methods cannot simulate actual passenger behavior on stairs under deeply buried conditions, nor can they reproduce passenger fatigue and the decrease in walking speed on long staircases, nor can they describe the differentiated choices passengers make between stairs, escalators, and elevators. This application's embodiments construct a pedestrian staircase walking speed decay model and a pedestrian choice model between stairs, escalators, and elevators, and integrate these models into a pedestrian dynamic simulation system. This allows the simulation results to more realistically reflect passenger decision-making and traffic characteristics under actual conditions, thereby significantly improving the accuracy and reliability of entrance and exit facility design.

[0039] Please see Figure 1 The illustration shows a flowchart of a method for configuring elevator facilities at the entrances and exits of a deeply buried subway station, as provided in an embodiment of this application. This method can be applied to electronic devices, which may include physical devices such as servers, PCs, tablets, or smartphones, or virtual devices such as virtual machines or containers. The electronic device can be a single device, a combination of multiple devices, or a cluster of numerous devices. The method for configuring elevator facilities at the entrances and exits of a deeply buried subway station may include:

[0040] Step S110: Obtain the design input information for the target subway station, which includes predicted passenger flow data and station burial depth data.

[0041] Step S120: Based on the design input information, construct a pedestrian dynamic simulation model; the pedestrian dynamic simulation model includes a fatigue model and a facility selection model; the fatigue model is used to simulate the speed decay of passengers walking on long staircases; the facility selection model is used to simulate the path selection of passengers between stairs, escalators and elevators.

[0042] Step S130: Obtain the pre-set initial facility configuration parameters and input the facility configuration parameters into the pedestrian dynamic simulation model for simulation operation; the initial facility configuration parameters include the specifications and quantity parameters of stairs, escalators and elevators.

[0043] Step S140: Based on the pedestrian spatiotemporal data collected during the simulation, calculate multiple performance indicators corresponding to the initial facility configuration parameters; these performance indicators include passage time, queuing time, congestion density, and fatigue level.

[0044] Step S150: Using multiple performance indicators as optimization objectives and initial facility configuration parameters as decision variables, a multi-objective optimization algorithm is adopted to generate the entrance and exit elevator facility configuration parameters of the target subway station.

[0045] In step S110, the predicted passenger flow data refers to data obtained through historical statistics, traffic model predictions, or urban planning data, reflecting the passenger flow and movement characteristics that the target subway station may carry during future operating hours. Specifically, it may include passenger flow in and out of the station at different times, the proportion of passenger flow directions, and passenger flow composition. This data can come from urban transportation planning reports, operational data of similar stations, or professional passenger flow prediction software.

[0046] Station depth data refers to the vertical difference in elevation between the subway station concourse and the ground-level entrances / exits. This geographical and structural parameter determines the physical height that passengers need to overcome to complete vertical passage, and its value can be directly obtained from the station's architectural structural design drawings or geological survey reports. As one implementation method, this information can be collected and formatted automatically or semi-automatically by reading design document databases, manually inputting design parameter files, or interacting with a BIM (Building Information Modeling) system, providing structured input for subsequent simulations.

[0047] In step S120, the pedestrian dynamic simulation model is an integrated system with a social force model as its underlying computing engine. The social force model treats each passenger as a particle subjected to various "forces" (such as self-driving force, repulsive force from others, and obstacle force), calculates its movement speed and direction, and constitutes the physical basis of the simulation.

[0048] Based on the social force model, this application integrates a fatigue model and a facility selection model to form a pedestrian dynamic simulation model. The fatigue model is a mathematical function used to quantify the decline in walking speed caused by physical exertion when passengers are ascending or descending long staircases. The fatigue model is implemented by using station burial depth data as a key input, combined with physiological parameters such as basal metabolic rate, and dynamically calculating the real-time speed of each passenger through a speed decay function, for example, where speed decreases exponentially with walking time or height, thereby simulating the real fatigue effect of passengers walking slower and slower.

[0049] Facility selection models simulate passenger decision-making behavior when faced with stairs, escalators, and elevators. They are implemented by using passenger flow pressure reflected in predicted passenger flow data to construct a cost-based decision logic. The facility selection model calculates a generalized cost for each facility option, integrating local and global information. Local information includes current queue length and walking distance to the facility entrance; global information includes the estimated total time to reach the destination using the facility. Based on these costs, passengers choose a facility with a certain probability using a discrete selection model (such as a Logit regression model).

[0050] As one implementation method, the secondary development interface of the existing pedestrian simulation software platform or a self-developed simulation framework can be used to encode and implement the above-mentioned fatigue calculation module and selection decision module, and embed them into the main loop of the social force model. This allows passengers to update their speed according to their own fatigue state and dynamically adjust their path selection based on environmental information at each time step of the simulation.

[0051] In step S130, the initial facility configuration parameters are a set of values ​​defining the specific dimensions and quantities of all vertical transportation facilities at the entrance and exit. These parameters are a digital representation of the design scheme. The initial facility configuration parameters may include: the net width and total number of steps of stairs; the number of escalators, the width of each escalator, and their operating speed; the number of elevators, their rated passenger capacity, their operating speed, and their stopping arrangements, etc. The initial facility configuration parameters can be initially determined by the designer based on experience, or they can be randomly generated according to certain rules.

[0052] Simulation execution refers to inputting these parameters into the pedestrian dynamic simulation model constructed in step S120 and driving the model to execute the complete simulation process. For example, the simulation system generates 3D models of stairs, escalators, and elevators of corresponding sizes and locations in the virtual environment based on the parameters; then, based on predicted passenger flow data, it generates virtual passengers (Agents) at the entrance according to certain rules; finally, within the set simulation duration (e.g., simulating a 2-hour morning rush hour), it runs according to the model rules and records all process data. The simulation process can be completed on a computer by running a simulation program.

[0053] In step S140, pedestrian spatiotemporal data refers to the time and spatial trajectory information of each virtual passenger automatically recorded by the system throughout the simulation process. This includes timestamps of entering and leaving entrance / exit areas, start and end times of waiting at each facility (staircase, escalator, elevator lobby), times at different locations on passageways and stairs, local passenger density at those locations, and the cumulative height ascended or descended by walking. Multiple performance indicators are extracted from this raw spatiotemporal data and are core metrics used to evaluate the quality of the design scheme.

[0054] As one implementation method, a series of data acquisition probes and listeners are set up within the simulation program to continuously record the aforementioned data. After the simulation ends, a dedicated index calculation module is invoked to perform statistical analysis on the massive amount of raw data according to predefined formulas. For example: the travel time index is obtained by calculating the average of the "departure timestamp - entry timestamp" for all passengers; the queuing time index is obtained by averaging the waiting time of each passenger in all queues; the congestion density index is obtained by averaging the density experienced by passengers at various points in key areas (such as the middle of a staircase); and the fatigue level index is obtained by normalizing the cumulative ascent height of each passenger and then averaging it. These calculated index values ​​constitute the comprehensive performance report of the scheme corresponding to the initial facility configuration parameters.

[0055] In step S150, several performance indicators, such as passage time and queuing time, are formally defined as optimization objectives that need to be minimized. Initial facility configuration parameters, such as stair width and number of elevators, are defined as adjustable decision variables.

[0056] Multi-objective optimization algorithms (such as NSGA-III) are intelligent search algorithms that solve problems involving the simultaneous optimization of multiple objectives. Their implementation is an iterative loop: First, the algorithm randomly generates or generates a batch of different combinations of facility configuration parameters based on rules, forming an initial group of schemes (initial population). Then, for each set of parameters in the initial population, steps S130 and S140 are automatically repeated—that is, the parameters are input into a simulation model and run to calculate their corresponding multiple performance index values. Next, based on the index values ​​of all schemes and the construction cost, which is usually also considered, the optimization algorithm uses its core rules, such as non-dominated sorting, congestion calculation, and reference point association, to compare the merits of the schemes, select the best-performing schemes, and generate a new generation of schemes through simulated "evolution."

[0057] The above process is repeated cyclically. Ultimately, the algorithm outputs a Pareto optimal solution set, which contains a series of solutions from which final, quantified recommendations for entrance and exit elevator facility configuration parameters can be generated based on actual priorities (such as prioritizing queue reduction or cost control). Technically, the simulation evaluation program needs to be encapsulated as an objective function that the optimization algorithm can call, and the main loop logic of the optimization algorithm needs to be implemented using tools such as Python and MATLAB.

[0058] In the implementation of the above embodiments: by inputting actual data such as predicted passenger flow and station burial depth into a pedestrian dynamic simulation model that integrates fatigue and facility selection models, it is possible to dynamically simulate and evaluate any given facility configuration parameter scheme, obtaining multiple performance indicators reflecting the actual operation, such as travel time, queuing time, congestion density, and fatigue level. Furthermore, using a multi-objective optimization algorithm, with these performance indicators as objectives, the specifications and quantity parameters of stairs, escalators, and elevators are automatically searched and iteratively optimized on a large scale, ultimately generating a configuration scheme with better overall performance. This method overcomes the limitations of traditional designs relying on static specifications and human experience, transforming the configuration of facilities such as elevators from subjective judgment to an objective optimization process driven by simulation data. This significantly improves the scientific rigor, accuracy, and efficiency of deep-buried subway entrance and exit design, helping to better balance construction costs with passenger efficiency, safety, and comfort while meeting traffic needs.

[0059] Optionally, in this embodiment, the design input information also includes elevator capacity, which can be determined based on the size, speed, and number of people passing through per hour of the staircases and escalators.

[0060]

[0061] in For elevator access capacity, For the elevator's rated capacity, For the number of elevators, H d The distance for pedestrians to walk. For elevator speed, For the number of stops (e.g., subways are often two-level, two-station systems, so you can take...), ); For door opening / closing time and pre-start time; The time each person takes to go up and down the elevator; This is the ratio of passenger flow entering and exiting the station (considering two-way flow).

[0062] This application's embodiments are based on a social force model to dynamically simulate pedestrian behavior. Each pedestrian is modeled as an intelligent agent, dynamically adjusting their speed and direction based on the target point, environmental obstacles, and surrounding pedestrians in the simulation scenario. Pedestrian speed update equation:

[0063]

[0064] in Let i be the mass (kg) of pedestrian i. Let be the current speed of pedestrian i (m / s). For time, Let be the expected speed (m / s) of pedestrian i. Let be the unit vector representing the expected direction of pedestrian i. The relaxation time (the time constant for a pedestrian to adjust their speed to the desired speed, usually taken as 0.5~1.0s). The pedestrian-pedestrian interaction force is the sum of the repulsive and attractive forces between pedestrian i and other pedestrians j around them, used to avoid collisions or follow other pedestrians. The interaction force between pedestrian i and surrounding walls / obstacles, such as walls and pillars. The sum of the repulsive forces between them prevents pedestrians from penetrating the obstacle.

[0065] Based on the station burial depth data, the parameters of the fatigue model are determined. The fatigue model is a mathematical model used to quantitatively simulate the gradual decrease in walking speed due to physical exertion when passengers walk a long distance up stairs. Its parameters are the key variables in the model that determine the rate and magnitude of speed decay. First, it is necessary to establish the correlation between the station burial depth data and the total height the passenger needs to climb or the expected walking time. For example, the greater the burial depth, the longer the passenger needs to continuously walk on the stairs for the required time t or the equivalent distance.

[0066] As one implementation method, the fatigue model is as follows:

[0067]

[0068] in, For pedestrian speed, Let i be the expected speed of pedestrian i. The fatigue index. The distance traveled. For risk perception, For the perception of fatigue. This represents the minimum speed for pedestrians.

[0069] The fatigue index can be set according to needs, for example, 0.02 or 0.025; both risk perception and fatigue perception change dynamically, and When pedestrians feel that their current situation is high-risk, then ,on the contrary Our simulation process uses the usual methods. , This represents the minimum speed for pedestrians.

[0070] To quantify the pedestrian selection process among different facility combinations, this application proposes a pedestrian facility selection model. We quantify pedestrian facility choices and plan paths using a hierarchical approach, utilizing local information about the pedestrian's current area and global information about the buildings. This method assumes that pedestrians know the locations of all facilities on the platform before entering the facility and the queue situation in front of the facility (local information). It also assumes that pedestrians know the distances of different routes to their destination (global information). Using local and global information, the cost of using each entrance / exit facility is calculated. Based on the cost relationship, a logit regression selection model is established to set the probability of pedestrian selection.

[0071] Based on predicted passenger flow data, the weight parameters in the facility selection model are initialized. These weight parameters balance the importance of various cost factors (such as time, distance, and queuing) in the decision-making process. Predicted passenger flow data can be analyzed to estimate the potential service pressure on each facility. For example, if the prediction shows extremely high peak passenger flow, the weight coefficient of the "queuing factor" in the facility selection model can be increased in advance. This allows virtual passengers in the model to perceive queuing costs earlier and more sensitively, thus tending to choose facilities with shorter queues or make alternative route decisions earlier in the initial simulation phase. This is equivalent to injecting prior knowledge of passenger congestion into the model, making the simulated passenger decision-making behavior closer to the real psychology under high passenger flow pressure.

[0072] In this embodiment of the application, the facility selection model determines the probability of a pedestrian choosing a facility in the following way:

[0073] The generalized cost of pedestrians choosing each facility is calculated. The generalized cost is determined by local cost and global cost. Local cost includes the queuing factor determined by the number of people queuing in front of the facility and the distance factor for pedestrians to reach the facility entrance. Global cost is the estimated time required for pedestrians to reach their destination through the facility.

[0074] The calculation method for generalized cost is as follows:

[0075]

[0076] in For pedestrians, the generalized cost of facility i, For the local cost of facility i for pedestrians, Let i be the global cost for pedestrians to facility i.

[0077] The calculation method for local costs is as follows:

[0078]

[0079]

[0080] in, For the preference of facility i, Let be the distance preference factor for pedestrians regarding facility i. As a distance penalty for facility i, The distance the pedestrian currently travels to facility i. Let i be the queuing factor for pedestrians at facility i. For pedestrians choosing facility i, the queue waiting time is... A conversion factor is chosen for pedestrians queuing for facility i, typically set to 1. Let i be the local time factor for pedestrians with respect to facility i. The time it takes for a pedestrian to reach facility i.

[0081] The following is the method for calculating the global cost of facility i for pedestrians:

[0082]

[0083] in Let be the global time for the pedestrian to reach their destination via facility i, assuming there are no other people present. Let be the global time factor for pedestrians with respect to facility i.

[0084] Based on generalized cost, the probability of a pedestrian choosing each facility is calculated using a regression choice model. For example, the probability of a pedestrian choosing a facility can be calculated using the following formula:

[0085]

[0086] Among them, P i Let i be the probability of a pedestrian choosing facility i. Let i be the generalized cost to pedestrians for facility i.

[0087] A pedestrian dynamic simulation model is constructed based on the parameters of the fatigue model and the weight parameters in the facility selection model. Using the social force model as the basic framework, the parameterized fatigue model and facility selection model are integrated into the simulation loop of the social force model to form the pedestrian dynamic simulation model. At each simulation time step, the actual speed of the pedestrian is dynamically updated according to the fatigue model, and target facilities are assigned to the pedestrian according to the facility selection model.

[0088] The "construction" here refers to coupling and programming the configured behavioral model with a basic microscopic pedestrian motion simulation engine. For example, a mature social force model or cellular automata model can be chosen as the underlying core of the simulation, responsible for handling basic pedestrian movement, collision avoidance, and other physical behaviors. Then, the speed decay calculation module containing specific fatigue model parameters, determined in the first step, and the facility selection model decision logic module with initialized weight parameters in the second step, are written as independent software functions or agent behavior rules. Finally, these modules are integrated into each simulation time step loop of the basic simulation engine through programming. Specifically, at each time step, the system calls the facility selection model for each virtual passenger, calculating their choice based on the current environment (queue, distance) and preset weight parameters; then, based on whether they are climbing stairs and the time already traveled, it calls the fatigue model and updates their current speed according to its parameters; the underlying engine then drives their movement based on this speed and direction. Through this loop, a pedestrian dynamic simulation model that can simultaneously reflect deep-seated physical exertion and peak passenger flow decision-making behavior is successfully constructed, providing a reliable tool for subsequent scheme evaluation.

[0089] In the implementation of the above embodiments: First, the fatigue model is configured with targeted parameters using station burial depth data, enabling the model to accurately reflect the speed decay characteristics of passengers at different station depths. Simultaneously, the weight parameters of the facility selection model are initialized using predicted passenger flow data, ensuring that passenger decisions in the model align with behavioral patterns under actual passenger flow pressure. Finally, these two pre-configured core behavioral modules are integrated into the basic simulation engine to construct a highly customized pedestrian dynamic simulation model. This method ensures that the final simulation model is no longer general or idealized, but closely matches the specific burial depth conditions and passenger flow predictions of the target subway station, thus providing a highly reliable behavioral basis for the simulation evaluation of subsequent facility configuration schemes. The resulting optimized design scheme better meets the efficiency, safety, and comfort requirements of specific stations in real-world operating environments, improving the accuracy and reliability of the design.

[0090] To objectively evaluate the performance of various design schemes in the automated design cycle, this application constructs a performance evaluation module consisting of four characteristics: traffic efficiency, queuing experience, congestion risk, and pedestrian fatigue. During simulation, this module automatically records pedestrian spatiotemporal data and converts it into four core indicators: travel time, congestion density, queuing time, and fatigue level. All indicators are numerical outputs and can be directly used as input for multi-objective optimization algorithms.

[0091] Optionally, in this embodiment, multiple performance indicators include passage time, queuing time, congestion density, and fatigue level; based on pedestrian spatiotemporal data collected during simulation, multiple performance indicators corresponding to the initial facility configuration parameters are calculated, including:

[0092] Travel time is calculated based on pedestrian entry and exit timestamps. Travel time reflects the entire process from a passenger entering the entrance / exit space to completing vertical traffic flow; it is a core indicator for measuring overall efficiency. The calculation method for travel time is as follows:

[0093]

[0094] in Total number of employees; , The first The entry and exit timestamps of each pedestrian.

[0095] Queuing time is calculated based on the waiting time pedestrians spend at the entrances of various facilities. Queuing time reflects the congestion level at facility bottlenecks (such as stairwell entrances, elevator entrances, and escalator tops) and is an important basis for judging the rationality of the design. The calculation method for queuing time is as follows:

[0096]

[0097] Where Wait represents the queuing time. This indicates the cumulative waiting time for pedestrians in front of any facility. This represents the total number of employees.

[0098] Congestion density is calculated based on the instantaneous pedestrian density at preset key nodes. To assess the congestion risk at entrances and exits under peak passenger flow conditions, this study uses average congestion density as the core safety indicator. Pedestrians continuously perceive the surrounding local density when crossing key facility nodes such as stairs, escalators, and elevators. Therefore, the instantaneous density of each pedestrian at each node is recorded in the simulation, and the density experienced by the same pedestrian multiple times is averaged to finally calculate the overall average congestion density:

[0099]

[0100] Where MD is the crowding density. Total number of employees; No. The number of density samples experienced by pedestrians when crossing facility nodes; For the first The pedestrian was in the 1st Local density at the time of sampling (unit: ).

[0101] Fatigue levels are calculated based on the height changes of pedestrians during walking, after normalization. In deeply buried subway entrances and exits, passengers need to complete long vertical ascents or descents. Continuous upward / downward movement leads to significant physical exertion and causes phenomena such as decreased speed and increased rest. To quantify the impact of this process on pedestrian behavior, this study introduces a fatigue model into the simulation and constructs an average fatigue level index (Fatigue). The fatigue level of pedestrians is determined by the physical exertion they experience during passage, which is represented by the height changes of pedestrians during walking. After the simulation, the overall fatigue level is obtained by averaging the fatigue values ​​of all pedestrians, calculated as follows:

[0102]

[0103] Fatigue refers to the degree of fatigue. For pedestrians The fatigue level of pedestrians was recorded. To quantify this level of fatigue, we represent the change in height of pedestrians as a result of walking and then normalize the data.

[0104]

[0105] For pedestrians The degree of fatigue, The height at which pedestrians walk. These are the maximum and minimum walking heights for pedestrians, respectively. This parameter can be set according to actual conditions, for example... It can also be other values.

[0106] In the implementation of the above embodiments, key design evaluation dimensions such as traffic efficiency, queuing conditions, congestion risk, and fatigue levels are transformed into specific, repeatable, and quantifiable indicators based on clear data sources and calculation rules. For example, traffic time is calculated using timestamp differences, and queuing time is calculated using waiting time, which is intuitive and resistant to interference. By designing corresponding data collection and calculation paths for different phenomena, the evaluation of the design scheme is comprehensive and objective, covering different aspects from macro-efficiency to micro-experience. This standardized, multi-dimensional quantitative evaluation system provides a unified and reliable scale and basis for fairly comparing the merits of different facility configuration schemes.

[0107] Optionally, in this embodiment of the application, an NSGA-III algorithm suitable for subway entrance / exit simulation is proposed, which is implemented as follows: using multiple performance indicators as optimization objectives and initial facility configuration parameters as decision variables, a multi-objective optimization algorithm is adopted to generate the elevator facility configuration parameters for the entrance / exit of the target subway station, including:

[0108] An initial population containing decision variables is obtained, where each individual represents a set of initial facility configuration parameters. Decision variables refer to key parameters of entrance / exit facilities that can be adjusted and optimized during the optimization process, specifically including the total width of stairs, the total width and speed of escalators, the number of elevators and their rated speed, etc. The initial population is a term in multi-objective optimization algorithms (such as NSGA-III), referring to a set of multiple potential solutions randomly generated or generated according to certain rules at the start of optimization. Each individual is a member of this set, specifically referring to a complete and specific set of initial facility configuration parameters. For example, "individual A" might represent a set of parameter values ​​such as [staircase width 0.6 meters, escalator width 2.4 meters, 2 elevators, speed 1.5 meters / second, capacity 13 people]. Technically, a reasonable range of values ​​is usually set for each decision variable according to design specifications (such as minimum staircase width and elevator speed range). Then, using the computer's random number generation function, each variable is randomly assigned a value within this range, thus creating dozens to hundreds of different parameter sets. The set of these parameter sets constitutes the initial population.

[0109] For each individual in the initial population, a pedestrian dynamic simulation model is invoked to perform simulations and calculate corresponding performance indicators. Based on these performance indicators and facility construction costs, the initial population is non-dominated and ranked, and a new population is generated based on the ranking results. Invoking the pedestrian dynamic simulation model means taking the specific set of facility configuration parameters represented by a particular individual in the initial population as input conditions and running the pre-built pedestrian dynamic simulation program once.

[0110] For example, this can be achieved through an interface between automated scripts or optimization algorithm platforms and simulation software. The algorithm program reads each individual in the population, converts its parameters into a format recognizable by the simulation model, and then starts and controls the simulation program to execute a complete simulation. After the simulation, the program automatically collects pedestrian spatiotemporal data and calculates multiple performance indicators corresponding to the scheme according to predetermined formulas, such as average travel time, average queuing time, average congestion density, average fatigue level, and the estimated facility construction cost based on the parameters. The output of this step is to assign a clear and quantifiable set of performance and cost scores to each individual in the population.

[0111] The new population is used as the initial population for iterative execution until the termination condition is met, generating a Pareto optimal solution set. Non-dominated sorting is a hierarchical method in multi-objective optimization, used to compare the merits of solutions across multiple objectives. If a solution is no worse than another solution in all objectives and is better in at least one objective, then that solution dominates the other. Through pairwise comparisons, the entire population can be divided into different levels of non-dominated fronts, with the first front being the set of solutions not dominated by any other solution, and so on. Generating a new population refers to producing the next generation of solutions based on the sorting results, through operations simulating biological evolution. Technically, the algorithm first performs non-dominated sorting based on the indicators and cost values ​​of all individuals, selecting individuals with superior performance. Then, selection, crossover, and mutation operations are performed on these superior individuals: selection involves choosing parents from the superior individuals according to certain rules; crossover involves exchanging and combining some decision variable values ​​of two parent individuals to produce new offspring individuals; mutation involves slightly randomizing some decision variable values ​​of the offspring individuals. Through these operations, a new population that inherits excellent characteristics while possessing novelty is generated.

[0112] Iterative execution refers to using the generation of a new population in the third step as the starting point for the next round of evaluation and optimization, and repeatedly performing simulation evaluation, ranking, and new population generation. The termination condition is a preset loop stopping criterion, which commonly includes reaching the maximum number of iterations or the optimal frontier of the population no longer showing significant improvement after several consecutive generations.

[0113] The Pareto optimal set is the set of all individuals in the current population that belong to the first non-dominated frontier when the iteration meets the termination condition. The solutions in this set are "non-dominated" to each other, meaning that it is impossible to improve one objective without harming others. They represent the best trade-off boundaries achievable between different design objectives.

[0114] The main loop of the optimization algorithm will continue to iterate until the termination condition is triggered. Then the program outputs the Pareto optimal solution set, which contains multiple sets of optimal compromise facility configuration parameters and their corresponding index values.

[0115] From the Pareto optimal solution set, the configuration parameters of the entrance and exit elevator facilities for the target subway station are determined. For example, if the project budget is very tight, the solution with the lowest construction cost in the solution set can be selected; if passenger experience is more important, the solution with relatively shorter travel time and less fatigue can be selected. Through this data-driven decision-making, the final set of specific staircase, escalator, and elevator specifications and quantities is a customized and optimized configuration parameter for the entrance and exit elevator facilities of the target subway station, which can be directly used to guide subsequent detailed design and construction.

[0116] In the implementation of the above embodiments: First, a large number of different initial facility configuration parameter combinations are generated as an initial population. Then, a pedestrian dynamic simulation model is automatically invoked to simulate each scheme and calculate multiple target indicators such as passage time, queuing time, congestion density, fatigue level, and cost. Next, based on these indicators, non-dominated sorting is used to classify and screen the schemes, and an evolutionary operation is used to generate a new population of schemes with better performance. This process is repeated iteratively until a series of Pareto optimal schemes that achieve the best balance between multiple performance and cost objectives are automatically found for selection. This process transforms the originally experience-dependent and cumbersome manual calculation and scheme comparison work into a systematic automatic search and optimization process driven by computer simulation and intelligent algorithms. This improves the design efficiency and scientific nature of elevator facility configuration at deep-buried subway entrances and exits, and helps to better ensure traffic efficiency and passenger comfort while controlling construction costs.

[0117] In an optional embodiment, the decision vector X can be set as: Among them, w s w is the width of the staircase e v is the width of the escalator. e n represents the escalator's operating speed. l c represents the number of elevators. l v is the rated capacity of the elevator. l This refers to the elevator's operating speed.

[0118] According to relevant standards, the width of a two-way mixed-traffic staircase must not be less than 2.4m, thus there are staircase constraints: ;

[0119] Escalator parameter constraints: ;

[0120] The elevator constraints are: ;

[0121] The facilities are also designed to meet the minimum required capacity. ;

[0122] in To meet the minimum design standards expected to accommodate peak passenger flow, we introduce a flexible design space to reflect design costs. .

[0123] Normalization process: ;

[0124] Optimization goal:

[0125] ;

[0126] NSGA-III Algorithm:

[0127] In the automated design framework of this application embodiment, the NSGA-III algorithm uses facility configuration parameters as core decision variables to carry out the optimization process. The system first constructs a decision vector from design parameters such as stair width, escalator width, number of elevators, and elevator operating speed, and sets their value range and feasibility constraints according to relevant national standards and engineering conditions. Based on this, the algorithm generates a diverse initial population, where each individual corresponds to a complete subway entrance / exit facility configuration scheme.

[0128] Subsequently, NSGA-III invoked the pedestrian dynamic simulation model constructed in this application for each individual in the population. This simulation model comprehensively considers factors such as pedestrian speed decay due to fatigue on long staircases, passenger selection behavior among stairs, escalators, and elevators, and simulates the dynamic evolution of passenger flow under actual operating conditions, thereby obtaining multi-dimensional performance indicators such as travel time, queuing time, personnel density, fatigue index, and facility construction cost for the corresponding scheme. Due to the dynamic nature and behavioral heterogeneity of the simulation process, its output can reflect the real operating state, avoiding the problem that traditional static empirical values ​​cannot accurately evaluate the performance of the scheme.

[0129] After calculating the indices, NSGA-III uses the aforementioned multidimensional simulation results as the objective function value for multi-objective optimization, performing a non-dominated sort on the merged parent and offspring populations. Simultaneously, the algorithm employs a reference-point-driven selection mechanism to maintain the uniformity of the objective space distribution, enabling the optimization process to consider different objective directions. During iterative updates, the algorithm continuously identifies schemes with superior performance or better balance among multiple indices, gradually bringing the population closer to the true Pareto front.

[0130] In each generation's selection phase, NSGA-III uses a reference point association method to map individuals in the current frontier to a preset reference direction. It then selects representative solutions based on the vertical distance from each individual to the reference point, ensuring that the optimization results cover different design trade-offs, including prioritizing cost reduction, prioritizing improved traffic efficiency, or achieving a balance between efficiency and comfort. Ultimately, the algorithm outputs not a single solution, but a set of facility configuration schemes that approximate the Pareto optimal surface. Designers can then select the final design scheme that best suits their practical application, taking into account factors such as site depth, budget constraints, and passenger flow characteristics.

[0131] The beneficial effects of this application's embodiments include: First, the present invention constructs a dynamic pedestrian simulation model that can realistically reflect the passage characteristics of deeply buried subway entrances and exits, which is the most important technical foundation of the present invention. In this model, not only is the fatigue speed decay law of pedestrians traveling long distances up or down established, but also a pedestrian selection behavior model based on passage cost and congestion conditions among stairs, escalators, and elevators is constructed. This gives the entire simulation system dynamic, heterogeneous, and behavior-driven characteristics, enabling it to realistically reproduce the passage status of deeply buried entrances and exits. This simulation capability is unattainable by traditional static empirical methods and is the fundamental innovation of this invention.

[0132] 2. Secondly, this invention structures the key design parameters involved in entrance and exit facilities into an optimizable decision variable system, and establishes feasible constraints on these variables that conform to engineering specifications and actual construction conditions. By quantifying parameters such as stair width, escalator width, number of elevators, and elevator speed, which were originally determined by the designer's experience, this invention can transform the originally subjective facility configuration problem into a calculable and optimizable parameter space. This structuring process enables facility configuration to be optimizable for the first time, which is a prerequisite for the algorithm to perform automatic design and is also the second key technical point of this invention.

[0133] 3. Finally, this invention uses the multi-dimensional operational indicators obtained through dynamic simulation as the objective function, and introduces the NSGA-III multi-objective optimization algorithm to automatically optimize the facility configuration scheme. A reference point mechanism is used to maintain a uniform distribution in the objective space, ensuring that the optimization results not only approximate the Pareto optimal solution set but also cover different design preference directions. Through continuous algorithm iteration and scheme replacement, this invention can generate a set of optimal configuration schemes that balance multiple objectives such as traffic efficiency, queuing time, congestion density, fatigue level, and construction cost. This step realizes the transformation from "single static verification" to "multi-objective automatic optimization," which is the core innovation of this invention at the application level.

[0134] In summary, the key technical points of this invention are provided by the data foundation of "pedestrian dynamic simulation model with realistic behavior mechanism", the optimization space of "structured facility parameters and their feasibility constraint system", and the automatic design capability of "multi-objective optimization algorithm based on NSGA-III". The three form a complete technical closed loop, which together realizes the objectification, quantification and intelligent design of subway deep-buried entrance and exit facilities.

[0135] This application provides a device for configuring elevator facilities at the entrance and exit of a deeply buried subway station, including:

[0136] The information acquisition module is used to acquire the design input information of the target subway station, which includes predicted passenger flow data and station burial depth data.

[0137] The simulation model module is used to build a pedestrian dynamic simulation model based on the design input information. The pedestrian dynamic simulation model includes a fatigue model and a facility selection model. The fatigue model is used to simulate the speed decay of passengers when walking on long stairs. The facility selection model is used to simulate the path selection of passengers between stairs, escalators and elevators.

[0138] The simulation operation module is used to obtain the pre-set initial facility configuration parameters and input the facility configuration parameters into the pedestrian dynamic simulation model for simulation operation; the initial facility configuration parameters include the specifications and quantity parameters of stairs, escalators and elevators;

[0139] The performance index module is used to calculate multiple performance indicators corresponding to the initial facility configuration parameters based on the pedestrian spatiotemporal data collected during the simulation. These performance indicators include passage time, queuing time, congestion density, and fatigue level.

[0140] The optimization module is used to generate the configuration parameters of the entrance and exit elevator facilities of the target subway station by using multiple performance indicators as optimization objectives, initial facility configuration parameters as decision variables, and employing a multi-objective optimization algorithm.

[0141] Optionally, in this embodiment of the application, the fatigue model for the deeply buried subway entrance elevator facility configuration device is as follows:

[0142]

[0143] in, For pedestrian speed, Let i be the expected speed of pedestrian i. The fatigue index. For risk perception, For the perception of fatigue. This represents the minimum speed for pedestrians.

[0144] Optionally, in this embodiment of the application, the deeply buried subway entrance elevator facility configuration device determines the pedestrian's choice probability of a facility through the facility selection model in the following way: calculating the generalized cost of a pedestrian choosing each facility, the generalized cost being determined by local cost and global cost; the local cost includes a queuing factor determined by the number of people queuing in front of the facility, and a distance factor for the pedestrian to reach the facility entrance; the global cost is the estimated time required for the pedestrian to reach their destination through the facility; based on the generalized cost, the probability of the pedestrian choosing each facility is calculated through a regression selection model.

[0145] Optionally, in this embodiment of the application, the deeply buried subway entrance elevator facility configuration device includes multiple performance indicators such as travel time, queuing time, congestion density, and fatigue level; the performance indicator module is specifically used to calculate the travel time based on the pedestrian's entry and exit timestamps; calculate the queuing time based on the pedestrian's waiting time in front of each facility entrance; calculate the congestion density based on the instantaneous density of pedestrians at preset key nodes; and calculate the fatigue level based on the height data of pedestrians changing through walking, after normalization processing.

[0146] Optionally, in this embodiment of the application, the deeply buried subway entrance elevator facility configuration device includes an optimization module, which is used to obtain an initial population containing decision variables, where each individual in the initial population represents a set of initial facility configuration parameters; for each individual in the initial population, a pedestrian dynamic simulation model is invoked to perform simulation and calculate the corresponding multiple performance indicators; based on the multiple performance indicators and facility construction costs, the initial population is non-dominatedly sorted, and a new population is generated based on the sorting results; the new population is used as the initial population for iterative execution until the termination condition is met, generating a Pareto optimal solution set; and the entrance elevator facility configuration parameters of the target subway station are determined from the Pareto optimal solution set.

[0147] Optionally, in the embodiments of this application, the decision variables for the deeply buried subway entrance elevator facility configuration device include: staircase width, escalator width, number of elevators, and rated elevator speed.

[0148] Optionally, in this embodiment of the application, the deep-buried subway entrance elevator facility configuration device and simulation model module are specifically used to determine the parameters of the fatigue model based on the station burial depth data; initialize the weight parameters in the facility selection model based on the predicted passenger flow data; and construct a pedestrian dynamic simulation model according to the parameters of the fatigue model and the weight parameters in the facility selection model.

[0149] It should be understood that this device corresponds to the above-described embodiment of the method for configuring elevator facilities at the entrances and exits of deeply buried subway stations, and is capable of performing the various steps involved in the above-described method embodiments. The specific functions of this device can be found in the description above, and detailed descriptions are omitted here to avoid repetition. The device includes at least one software functional module that can be stored in a memory or embedded in the device's operating system (OS) in the form of software or firmware.

[0150] Please see Figure 2 The diagram shows a structural schematic of an electronic device provided in an embodiment of this application. An electronic device 300 provided in this application includes a processor 310 and a memory 320. The memory 320 stores machine-readable instructions executable by the processor 310. When the machine-readable instructions are executed by the processor 310, the method described above is performed.

[0151] Figure 2 The components shown can be implemented using hardware, software, or a combination thereof. Electronic device 300 may be a physical device, such as a server or PC, or a virtual device, such as a virtual machine or virtualization container. Furthermore, electronic device 300 is not limited to a single device; it can be a combination of multiple devices or a cluster of numerous devices.

[0152] This application also provides a storage medium storing a computer program, which is executed by a processor to perform the above-described method.

[0153] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0154] This application also provides a computer program product, including computer program instructions, which are executed by a processor to perform the method described above.

[0155] It should be understood that the disclosed apparatus and methods can also be implemented in other ways, given the several embodiments provided in this application. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0156] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0157] The above description is only an optional implementation of the embodiments of this application, but the protection scope of the embodiments of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of this application should be covered within the protection scope of the embodiments of this application.

Claims

1. A method for configuring elevator facilities at a deeply buried subway entrance, characterized in that, include: Obtain the design input information for the target subway station, including predicted passenger flow data and station burial depth data; Based on the design input information, a pedestrian dynamic simulation model is constructed; the pedestrian dynamic simulation model includes a fatigue model and a facility selection model; the fatigue model is used to simulate the speed decay of passengers walking on long staircases. The facility selection model is used to simulate the passenger's path selection among stairs, escalators, and elevators; Obtain the pre-set initial facility configuration parameters and input the facility configuration parameters into the pedestrian dynamic simulation model for simulation operation; the initial facility configuration parameters include the specifications and quantity parameters of stairs, escalators and elevators; Based on the spatiotemporal data of pedestrians collected during the simulation, multiple performance indicators corresponding to the initial facility configuration parameters are calculated; these multiple performance indicators include passage time, queuing time, congestion density, and fatigue level. Using the aforementioned multiple performance indicators as optimization objectives and the initial facility configuration parameters as decision variables, a multi-objective optimization algorithm is employed to generate the entrance and exit elevator facility configuration parameters for the target subway station. The fatigue model is as follows: in, For pedestrian speed, Let i be the expected speed of pedestrian i. The fatigue index. The distance traveled. For risk perception, For the perception of fatigue. Minimum speed for pedestrians; The facility selection model determines the probability of a pedestrian's choice of a facility by: The generalized cost of a pedestrian choosing each facility is calculated, and the generalized cost is determined by local cost and global cost; the local cost includes a queuing factor determined by the number of people queuing in front of the facility, and a distance factor for the pedestrian to reach the facility entrance; the global cost is the estimated time required for the pedestrian to reach their destination through the facility. Based on the generalized cost, the probability of pedestrians choosing each facility is calculated using a regression choice model.

2. The method according to claim 1, characterized in that, Based on pedestrian spatiotemporal data collected during the simulation, multiple performance indicators corresponding to the initial facility configuration parameters are calculated, including: The passage time is calculated based on the pedestrian's entry and exit timestamps; The queuing time is calculated based on the waiting time of pedestrians at the entrances of each facility; The congestion density is calculated based on the instantaneous pedestrian density at preset key nodes; The fatigue level is calculated based on the height data of pedestrians changing as they walk, after normalization processing.

3. The method according to claim 1, characterized in that, Using the aforementioned performance indicators as optimization objectives and the initial facility configuration parameters as decision variables, a multi-objective optimization algorithm is employed to generate the entrance and exit elevator facility configuration parameters for the target subway station, including: Obtain an initial population containing the decision variables, wherein each individual in the initial population represents a set of initial facility configuration parameters; For each individual in the initial population, the pedestrian dynamic simulation model is invoked to perform simulation and calculate the corresponding multiple performance indicators. Based on the aforementioned performance indicators and facility construction costs, the initial population is subjected to non-dominated sorting, and a new population is generated based on the sorting results. The new population is used as the initial population for iterative execution until the termination condition is met, generating a Pareto optimal solution set. From the Pareto optimal solution set, determine the configuration parameters of the elevator facilities at the entrances and exits of the target subway station.

4. The method according to claim 3, characterized in that, The decision variables include: staircase width, escalator width, number of elevators, and rated elevator speed.

5. The method according to any one of claims 1-4, characterized in that, The design input information also includes elevator access capacity. Based on the design input information, a pedestrian dynamic simulation model is constructed, including: Based on the site burial depth data, the parameters of the fatigue model are determined; Based on the predicted passenger flow data, initialize the weight parameters in the facility selection model; A pedestrian dynamic simulation model is constructed based on the parameters of the fatigue model and the weight parameters in the facility selection model.

6. A computer program product, characterized in that, It includes computer program instructions that are executed by a processor to perform the method as described in any one of claims 1 to 5.

7. An electronic device, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, perform the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, perform the method as described in any one of claims 1 to 5.

Images