Passenger evacuation method, device, equipment, storage medium and computer program product

By collecting and analyzing multi-source data and using traffic models to build passenger flow evacuation strategies, the problem of insufficient flexibility and accuracy of emergency passenger flow evacuation strategies in the existing technology is solved, and real-time prediction and dynamic adjustment of emergency passenger flow of rail transit is achieved. Improved evacuation efficiency.

CN118761886BActive Publication Date: 2024-11-12BEIJING URBAN ENG DESIGN & RES INST CO LT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411224215.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-11-12
Estimated Expiration
2044-09-03

AI Technical Summary

Technical Problem

There is a lack of systematic research on emergency passenger flow evacuation strategies in existing rail transit operations, and it is difficult to respond to and adapt to different emergencies in a timely manner. There is insufficient understanding of the daily rail transit network travel characteristics, resulting in low accuracy of emergency strategies.

Method used

By collecting multi-source data, including passenger travel data and mobile phone signaling data, calculate the OD path probability of the rail transit network and the number of people and jobs in the mobile phone signaling grid, use traffic models to build passenger flow evacuation strategies, and use advanced machine learning algorithms and traffic models achieve real-time prediction and dynamic adjustment.

Benefits of technology

It improves the targeted response and the flexibility of strategy, enhances the prediction and evacuation efficiency of rail transit emergency passenger flow, provides deeper insights into passenger flow behavior, and provides scientific basis for the formulation of emergency evacuation strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761886B_ABST
    Figure CN118761886B_ABST
Patent Text Reader

Abstract

The present invention discloses a passenger evacuation method, device, equipment, storage medium and computer program product, the method comprising: collecting passenger travel data and mobile phone signaling data; calculating the probability of each effective path in the OD path of the rail transit network based on the passenger travel data; calculating the number of permanent residents and the number of posts in the mobile phone signaling grid based on the mobile phone signaling data; based on the passenger travel data, the mobile phone signaling data, the number of permanent residents, the number of posts in the mobile phone signaling grid and the probability of each effective path in the OD path, obtaining a passenger evacuation strategy through a traffic model. Through the present invention, the technical problems existing in the prior art in terms of responding to emergencies, the flexibility of passenger evacuation strategies and the accuracy of emergency strategies are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of urban planning, and in particular relates to a passenger evacuation method, device, equipment, storage medium and computer program product. Background Art

[0002] At present, in the field of rail transit operations, emergency passenger evacuation is an important issue to ensure system safety and smooth travel of passengers.

[0003] Traditional methods are mainly based on management applications, tending towards administrative processes and response methods, but lacking systematic research on predicting the scale and direction of evacuated passenger flow, resulting in untimely emergency response; and passenger evacuation strategies are fixed and difficult to adapt to different emergency situations; at the same time, there is insufficient understanding of the travel characteristics of the daily rail transit network, resulting in low accuracy in the formulated emergency strategies. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a passenger evacuation method, device, equipment, storage medium and computer program product.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a passenger evacuation method, the method comprising:

[0007] Collecting multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data;

[0008] The probability of each valid path in the OD path of the rail transit network is calculated based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway;

[0009] Calculate the number of permanent residents and the number of jobs within the mobile phone signaling grid based on the mobile phone signaling data;

[0010] Based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, a passenger flow evacuation strategy is obtained through a traffic model.

[0011] In a second aspect, an embodiment of the present invention provides a passenger evacuation device, the device comprising:

[0012] A data collection module is configured to collect multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data;

[0013] A first calculation module is configured to calculate the probability of each valid path in the OD path of the rail transit network based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway;

[0014] A second calculation module is configured to calculate the number of permanent residents and the number of jobs in the mobile phone signaling grid based on the mobile phone signaling data;

[0015] The strategy determination module is configured to obtain a passenger flow evacuation strategy through a traffic model based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path.

[0016] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising: a memory, a processor; the processor is used to read and execute a computer program stored in the memory to implement the steps of the aforementioned passenger evacuation method.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and when the computer executable instructions are executed, the steps of the aforementioned passenger evacuation method are implemented.

[0018] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the aforementioned passenger evacuation method.

[0019] The beneficial effects brought by the technical solution provided by the embodiment of the present invention include:

[0020] Collect multi-source data, including passenger travel data and mobile phone signaling data; scientifically classify the currently monitored passenger evacuation event data based on historical data analysis and real-time prediction results, and formulate corresponding evacuation strategies. This classification method not only improves the pertinence of emergency response, but also enhances the flexibility and adaptability of the strategy; by integrating multiple data sources, a comprehensive rail transit passenger behavior analysis platform is built. This data integration method can provide deeper insights into passenger behavior and provide a scientific basis for the formulation of emergency evacuation strategies;

[0021] Based on the passenger travel data, the probability of each valid path in the OD path is calculated, wherein the OD path is the path between any departure station O and any target station D of the subway; based on the mobile phone signaling data, the number of permanent residents and the number of jobs in the mobile phone signaling grid are calculated; based on the multi-source data, the number of permanent residents and the number of jobs in the mobile phone signaling grid and the probability of each valid path in the OD path, a passenger flow evacuation strategy is obtained through a traffic model, so as to evacuate passengers based on the passenger flow evacuation strategy, and adopt advanced machine learning algorithms and traffic models to realize real-time prediction and dynamic adjustment of rail transit emergency passenger flow. This mechanism can ensure that in an emergency, the evacuation strategy can quickly adapt to the changing passenger flow demand and traffic conditions, and improve the evacuation efficiency. Through the present invention, the technical problems existing in the prior art in terms of responding to emergencies, the flexibility of passenger flow evacuation strategies and the accuracy of emergency strategies are solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A schematic diagram of a flow chart of an embodiment of a passenger evacuation method of the present invention;

[0024] Figure 2 for Figure 1 A detailed flow chart of step S40;

[0025] Figure 3 A schematic diagram of a flow chart of another embodiment of the passenger evacuation method of the present invention;

[0026] Figure 4 It is a functional module schematic diagram of an embodiment of a passenger evacuation device of the present invention;

[0027] Figure 5 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0030] In a first aspect, an embodiment of the present invention provides a passenger evacuation method.

[0031] In one embodiment, referring to Figure 1 , Figure 1 FIG. 1 is a flow chart of an embodiment of a passenger evacuation method of the present invention. Figure 1 As shown, passenger evacuation methods include:

[0032] Step S10, collecting multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data;

[0033] In this embodiment, the preset accident level is level 1 evacuation. When the accident level corresponding to the currently monitored passenger evacuation event data is level 1 evacuation, the evacuation mainly relies on the internal measures of the station, analyzes the passenger flow trend around the station, and optimizes the evacuation instructions and paths inside the station. When the accident level corresponding to the currently monitored passenger evacuation event data is greater than level 1 evacuation, the key operation data of the rail transit system is comprehensively collected through the data acquisition module to build a detailed passenger behavior and system status database.

[0034] The specific contents collected by the data collection module are described as follows:

[0035] (1) Rail transit passenger travel data collection

[0036] Through the automatic fare collection system (AFC) records, passengers' travel information is accurately counted, including the number of passengers entering and leaving each station, the time distribution of entry and exit, etc., and the number of passengers entering and leaving each station in each time period is accurately counted, providing a basis for passenger flow analysis and emergency evacuation of the rail transit system.

[0037] (2) Rail transit vehicle operation data collection

[0038] Collect existing train operation data, including departure time, arrival time and other information, to predict and adjust train operation plans, and quickly deploy resources in emergency situations.

[0039] (3) Event-related data collection

[0040] Through the historical event database, the impact of emergency events on the operation of the rail transit system is collected and analyzed, so that effective emergency evacuation strategies can be quickly formulated when similar events occur.

[0041] (4) External environment data collection

[0042] Monitor the impact of external factors such as weather changes and special events on the rail transit system to evaluate the impact of the external environment on passenger evacuation and formulate corresponding response measures.

[0043] (5) Multimodal traffic data collection

[0044] By integrating and analyzing data from other modes of transportation such as buses and taxis, comprehensive transportation network guidance is provided for the rail transit system, so as to provide passengers with a variety of evacuation options in case of emergency evacuation.

[0045] Considering the possible limitations of the rail transit network in an emergency, the present invention proposes an evacuation solution that combines multiple modes of transportation such as buses and taxis. This multi-modal integration strategy can provide passengers with more diverse evacuation options and ensure the smooth progress of the evacuation process.

[0046] (6) Data collection on the job distribution of the entire city’s population

[0047] Collect and analyze citywide population distribution, job distribution, and commuting patterns between population and jobs. Obtain census data, employment statistics, and urban planning information. In addition, use geographic information system (GIS) technology to combine this data with rail transit network maps to identify potential peak passenger flow areas.

[0048] (7) Travel characteristic data collection based on mobile phone signaling data

[0049] Analyze passengers’ travel patterns using mobile phone signaling data, including travel frequency, travel time, common routes, etc. Obtain anonymized mobile phone signaling data and infer passengers’ travel paths and behavior patterns through location data analysis technology. Combine these data with rail transit network maps using geographic information system (GIS) technology to reveal travel characteristics outside the rail transit system and provide support for multi-modal transportation integration.

[0050] In the initial stage of rail transit passenger travel route selection and passenger flow feature extraction, the data collected by the automatic ticket collection system (AFC) needs to be thoroughly cleaned. This process includes identifying and removing outliers as required, handling missing data, and ensuring data consistency. The data is then formatted to ensure that all data sets follow a unified standard format for subsequent analysis. Finally, normalization is performed to ensure that data of different magnitudes can be compared and analyzed on the same scale.

[0051] By integrating multiple data sources such as population position data, mobile phone signaling data, bus card swiping data, subway card swiping data, etc., a comprehensive rail transit passenger behavior analysis platform was built. This data integration method can provide deeper insights into passenger behavior and provide a scientific basis for the formulation of emergency evacuation strategies.

[0052] Furthermore, in terms of data collection and preprocessing, cloud computing and big data processing platforms can be considered, which will improve the efficiency of data processing and enhance the scalability of the system.

[0053] Step S20, calculating the probability of each valid path in the OD path of the rail transit network based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway;

[0054] In some specific embodiments, step S20 includes:

[0055] Selecting any path in the OD path of the rail transit network as a target path, preprocessing the passenger travel data corresponding to the target path, and obtaining the preprocessed passenger travel data;

[0056] Substitute the preprocessed passenger travel data into the first preset formula to calculate the path cost of the target path. The first preset formula is as follows:

[0057]

[0058] In the formula, represents the path cost of the target path, m represents the number of inter-station paths between two adjacent stations in the OD path, represents the i-th inter-station path in the OD path, represents the running time of the train on the i-th station path in the OD path, n represents the number of transfer station paths in the OD path, represents the transfer factor required by the passenger for the jth transfer path within the OD path, It represents the transfer time required by the passenger to take the jth transfer path within the OD path;

[0059] Select any path from the OD paths of the rail transit network that have not been selected as the target path, return to the step of preprocessing the passenger travel data corresponding to the target path to obtain the preprocessed passenger travel data, until all paths in the OD paths of the rail transit network have been selected, and obtain the path cost of each path in the OD paths of the rail transit network;

[0060] Determining a minimum path cost from the path costs of each of the paths;

[0061] Substitute the minimum path cost into the second preset formula to calculate the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. The second preset formula is as follows:

[0062]

[0063] In the formula, It represents the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. It represents the minimum path cost within the OD path of the rail transit network, a and b are constants, and min() is the minimum value function;

[0064] From the path costs of each path, a path cost that is less than or equal to the maximum path cost threshold is selected as a target path cost, and a path corresponding to the target path cost is taken as a valid path;

[0065] Based on the path costs of the effective paths, the probability of each effective path in the OD path of the rail transit network is calculated.

[0066] In this embodiment, the passenger travel data includes: the number of inter-station paths between two adjacent stations in the OD path, the number of transfer station paths in the OD path, the attractiveness correction factor of the i-th inter-station path in the OD path, the running time of the train on the i-th inter-station path in the OD path, the transfer factor required by the passenger for the j-th transfer path in the OD path, and the transfer time required by the passenger for the j-th transfer path in the OD path; the preprocessing includes: data filtering processing, formatting processing and normalization processing.

[0067] After data preprocessing, it is necessary to identify all possible travel paths that passengers may choose in the rail transit network. Through the path cost calculation method algorithm, the possible paths that passengers may take between different starting points and end points are determined. At the same time, factors such as the convenience of rail transfers will also be considered to screen out effective paths that passengers are more likely to choose in practice.

[0068] The main factors affecting passenger path selection include travel time, travel distance, number of transfers, train comfort, etc., and the most important factor is travel time. In order to uniformly measure these factors, other factors are uniformly converted into travel time, which is called path cost. The formula corresponding to the path cost is the first preset formula. Substituting the pre-processed passenger travel data corresponding to each path in the OD path of the rail transit network into the first preset formula, the path cost of each path in the OD path of the rail transit network can be calculated.

[0069] Sort the path cost of each path within the OD path, or determine the minimum path cost from the path cost of each path within the OD path through the min() function.

[0070] Substituting the minimum path cost into the second preset formula, the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network can be calculated: From the path cost of each path in the OD path of the rail transit network, the path cost that is less than or equal to the maximum path cost threshold is selected as the target path cost, and the path corresponding to the target path cost is taken as the valid path.

[0071] Furthermore, for the identification of effective rail transit paths, we can explore the use of graph neural networks (GNNs) to identify and optimize passengers' travel paths. This method can better handle complex rail transit networks. In addition, heuristic algorithms such as genetic algorithms can also be used to find the optimal path to adapt to the changing passenger flow demand.

[0072] Based on the path cost of the effective path, the probability of each effective path in the OD path of the rail transit network is calculated. The passenger flow allocation model is applied to estimate the probability of passengers choosing between different paths and calculate the passenger flow distribution of each path. The Logit model is a commonly used multi-path selection probability model that can predict the path choice of passengers based on their travel costs (such as travel time, number of transfers, etc.).

[0073] Specifically, for the probability of the kth valid path, the path cost of the kth valid path in the OD path and the minimum path cost in the OD path are substituted into the third preset formula to calculate the probability of the kth valid path in the OD path of the rail transit network. The third preset formula is as follows:

[0074]

[0075] In the formula, represents the probability of the kth valid path in the OD path of the rail transit network, represents the corrected path cost of the kth effective path in the OD path of the rail transit network, , represents the normal parameter, , Represents the path cost of the kth valid path in the OD path of the rail transit network Minimum path cost within the OD path of the rail transit network The ratio of K represents the number of valid paths within the OD path of the rail transit network.

[0076] Since the value space of the right half of the normal function is [1, +∞), in order to meet the range requirement, the path cost is first transformed into a ratio to the minimum path cost, and then the value is input into the function to obtain the corrected path cost; the normal parameter affects the steepness of the normal curve, and the value range is (0, +∞), which is called "path selection sensitivity". Its physical meaning is: as the path cost increases, the speed at which the passenger's rejection of the path increases, and its value is inversely proportional to the speed. The probability calculated with the corrected path cost as the value, its physical meaning is: the probability of a passenger choosing to take an effective path between OD.

[0077] Step S30, calculating the number of permanent residents and the number of jobs in the mobile phone signaling grid based on the mobile phone signaling data;

[0078] In some specific embodiments, step S30 includes:

[0079] Perform spatial data fusion and sampling expansion operations on mobile phone signaling data and permanent population numbers to obtain the permanent population numbers within the mobile phone signaling grid;

[0080] The mobile phone signaling data and the number of positions are spatially fused and expanded to obtain the number of positions in the mobile phone signaling grid.

[0081] In this embodiment, the mobile phone signaling data is used to determine the user's spatial location through information exchange between mobile phone users and base stations.

[0082] By expanding and verifying the mobile phone signaling data in combination with population and job data, the characteristics of urban travel are comprehensively obtained, including the occurrence volume, attraction volume, and travel time and space distribution. This process aims to provide a more comprehensive and accurate data set for in-depth analysis of urban traffic characteristics, and thus provide a scientific basis for multi-mode evacuation of road traffic emergency passenger flow.

[0083] First of all, it is necessary to divide the city into detailed units. By dividing the urban geographic space according to the planned and current urban functional areas, ensure that each unit has unique functional characteristics and travel characteristics. The ideal area range of urban units should be between 1 square kilometer and 10 square kilometers to facilitate subsequent data analysis and management. At the same time, it is necessary to combine the rail transit network to identify and mark the urban functional areas where each station is located, so as to integrate and analyze the city's travel data with the rail transit travel data. By dividing the city into detailed units and combining it with the rail transit network, the city's travel characteristics are deeply analyzed. This method helps to identify potential passenger flow peak areas and provide accurate geographical and spatiotemporal information for the formulation of evacuation strategies.

[0084] Then, the mobile phone signaling data will be integrated with the population and job survey data. By integrating the two data sources and expanding the sample, the deficiencies of the original data will be compensated and its representativeness will be improved, so that accurate data analysis can be carried out at the urban unit scale. In the process of expanding the sample, factors such as population density, employment distribution, and the relationship between residence and work should be considered to ensure that the expanded data can more comprehensively reflect the travel characteristics of urban residents.

[0085] The weighted average method and other methods can be used to expand the sample of mobile phone signaling data. Weighted average method example:

[0086] Weighted average is a common data processing method in statistics. It can be used to calculate the average of a set of data while taking into account the importance or weight of each data point. In the process of expanding the sample, the weight of each data point can be determined based on factors such as its representativeness in the population or the credibility of the data point. The following are the basic steps for expanding the sample using weighted average:

[0087] First, a weight needs to be assigned to each urban unit. The weight is usually related to the representativeness of the population. For example, if the demographic characteristics of an area (such as age, gender, occupation, etc.) are similar to those of the overall population, then the weight of this area may be higher. Weights can also be determined based on factors such as population density, land use type, etc.

[0088] The sampling expansion rate is then calculated, which reflects the user representativeness and can be calculated by comparing the number of permanent mobile phone users in a certain spatial unit with the number of permanent residents in the statistical data.

[0089] Finally, the weighted average method is used to calculate the estimated value of the number of permanent residents or the number of jobs. The value of each data point is multiplied by its corresponding weight, then all the products are added together and finally divided by the sum of the weights. The calculation of the number of permanent residents in the mobile signaling grid is as follows:

[0090]

[0091] In the formula, is the number of permanent residents in the mobile signaling grid; It is the first The resident population of the data point; It is the first The population density weight of the data point.

[0092] The number of positions within the mobile signaling grid is calculated as follows:

[0093]

[0094] In the formula, is the number of posts in the mobile signaling grid; It is the first The number of positions at the data point; It is the first The position weight of the data point.

[0095] Detailed data of key urban units are collected through traffic surveys and other means, including information on residents' travel time and space distribution characteristics, transportation mode selection, etc., as a benchmark for verifying the expanded sample results. This data is then used to verify the expanded sample data, and the weighted average method is used for iterative calculation to improve the accuracy and reliability of the data.

[0096] Step S40, based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, a passenger flow evacuation strategy is obtained through a traffic model.

[0097] The traffic model includes: a travel generation model, a travel distribution model, a traffic mode division model and a traffic assignment model. Figure 2 , Figure 2 for Figure 1 Detailed flow chart of step S40 in FIG. Figure 2 As shown, step S40 includes:

[0098] Step S401, based on the number of permanent residents and the number of jobs in the mobile phone signaling grid, the total number of travelers in cell w and the number of jobs in cell w are calculated through a travel generation model. Total population entering

[0099] In this embodiment, the travel generation model predicts the travel volume that residents may generate at the two endpoints of travel generation and attraction under certain socio-economic development conditions and a specific land use layout. Due to the imbalance of regional development during urban development, the intensity of residents' travel volume in different regions will vary to a certain extent. The travel generation model includes a travel generation model and a travel attraction model.

[0100] The travel generation model calculates the travel data (mainly the travel rate) of various travel purposes of each traffic community in one day according to the daily travel activities (or travel rate) of residents in the traffic community, and calculates the four purposes of each community, namely, home-based work, home-based school, home-based other travel, and non-home-based travel, according to the number of jobs provided in the community, such as employment, schooling, medical treatment, leisure, and shopping. In the present invention, the generation of travel is divided into two types, the first is the existing background travel generation in the model, and the second is the travel generation that requires evacuation of passenger flow. The travel generation model is used to obtain the travel rate of the qth travel purpose and the number of permanent residents of the qth travel purpose in the number of permanent residents in the mobile phone signaling grid where the community w is located. Substitute the travel rate of the qth travel purpose and the number of permanent residents of the qth travel purpose in the number of permanent residents in the mobile phone signaling grid where the community w is located into the fourth preset formula to calculate the total number of travel population in the community w, where the fourth preset formula is the travel generation model.

[0101] The attraction of different travel purposes is closely related to the characteristics of urban land use and jobs. Traffic communities with different locations and different traffic accessibility have significantly different travel attraction with the same land use nature and the same number of jobs. The attraction rate of the qth type of travel purpose and the number of jobs provided by the qth type of travel purpose in the number of jobs within the mobile phone signaling grid. The attraction rate of the qth type of travel purpose and the number of jobs provided by the qth type of travel purpose in the number of jobs in the mobile phone signaling grid are substituted into the fifth preset formula to calculate the cell The total inflow population, among which the fifth preset formula is the travel attraction model.

[0102] Among them, the fourth preset formula is as follows:

[0103]

[0104] Where: is the total number of people traveling in community w; is the travel rate of the qth travel purpose among the permanent population in the mobile phone signaling grid where cell w is located; is the number of permanent residents with the qth type of travel purpose among the permanent residents in the mobile phone signaling grid where cell w is located;

[0105] The fifth preset formula is as follows:

[0106]

[0107] Where: For the community Total population entering For the community The attraction rate of the qth type of travel purpose among the number of posts in the mobile signaling grid; For the community The number of positions provided by the qth travel purpose among the number of positions within the mobile phone signaling grid.

[0108] Step S402: based on the total number of people traveling in community w and the community The total population entering the community is calculated through the travel distribution model. Number of trips;

[0109] In this embodiment, the gravity model is used for travel distribution. The gravity model is similar to Newton's law of gravity, which holds that the travel volume between two communities is directly proportional to the attraction between the two communities and inversely proportional to the function of the spatial distance between the two communities. The gravity model takes into account the travel impedance between communities and can therefore be used to evaluate the effects of different land use / transportation system and network changes. The total population entering the community is substituted into the travel distribution model to calculate the distance from community w to community The number of trips, the travel distribution model is as follows:

[0110]

[0111] In the formula, From cell w to cell The number of trips, is the total number of people traveling in community w, For the community The total population entering is the impedance function, , From cell w to cell resistance, , and is a constant, is the total number of cells.

[0112] Step S403, based on the multi-source data and the The number of trips from community w to community w is calculated by dividing the model by traffic mode. Transportation of passenger traffic;

[0113] In some specific embodiments, step S403 includes:

[0114] Based on the multi-source data, the traffic mode classification model is used to calculate the distance from cell w to cell w. Transportation probability;

[0115] Using from cell w to cell Transportation The probability of cell w to cell The number of trips from cell w to cell Transportation of travel passenger flow.

[0116] In this embodiment, a traffic zone refers to a combination of areas with similar land use properties, resident composition, and traffic characteristics, and is a basic statistical analysis unit for predicting urban population size and distribution, social and economic development, land use planning, and urban traffic demand. After a city is divided into detailed units, multiple traffic zones can be obtained.

[0117] The mode division model adopts the Logit selection model, which is in the form of:

[0118]

[0119] Where: From cell w to cell Transportation The probability of, in this embodiment, may be the transportation mode to be selected by the people to be evacuated; is the total number of cells; For transportation From cell w to cell comprehensive costs.

[0120] The function form of the comprehensive cost is:

[0121] Where: For transportation From cell w to cell travel time; For transportation When leaving the community and entering the community time; For transportation From cell w to cell distance; For transportation From cell w to cell The monetary cost refers to the money that travelers must pay for riding motorized vehicles. The travel cost of public transportation such as buses and taxis is the fare, and the travel cost of cars includes fuel costs, road tolls, etc. For transportation From cell w to cell Other expenses, such as parking fees; , , , , , and The classification of traffic modes in this embodiment is mainly based on traffic evacuation strategies, and the actual parameters of the Logit model need to be determined according to actual conditions.

[0122] Calculate the distance from cell w to cell Transportation Probability After that, we can multiply the distance from cell w to cell Number of trips , get the distance from cell w to cell Transportation of passenger traffic , .

[0123] Step S404, based on the Transportation The passenger flow and the probability of each valid path in the OD path are used to obtain the traffic flow of each valid path through the traffic assignment model;

[0124] In this embodiment, the road network traffic distribution model is actually a path selection model, the purpose of which is to distribute the travel OD volume of each mode of transportation that needs to evacuate people to the paths of the urban transportation network. The input data of the traffic distribution model is the OD matrix of each travel mode obtained by the travel mode selection model.

[0125] First, based on the OD matrix and traffic mode selection results obtained in the traffic distribution stage, it is clear that Transportation of passenger traffic Next, the topological structure of the road network within the evacuation range is constructed using traffic model software, including road sections, nodes and their connection relationships. On this basis, an impedance function is defined for each road section according to the actual operation of urban traffic, usually expressed in travel time or cost, and reflects the impact of traffic on impedance.

[0126] In the traffic model, an appropriate allocation algorithm is selected according to user needs, and the shortest path of each OD pair is calculated using the traffic model software. Assigned to these paths, the traffic flow of each road segment is obtained.

[0127] Furthermore, in terms of passenger flow forecasting, deep learning techniques, such as convolutional neural networks (CNN) or recurrent neural networks (RNN), can also provide forecasts of passenger flow size and travel direction. Combining time series analysis and complex network theory, passenger flow dynamics can be more comprehensively understood and predicted.

[0128] Step S405, optimizing the passenger evacuation strategy based on the traffic flow of each effective path.

[0129] In this embodiment, after obtaining the allocation result, the evacuation effect after the implementation of the current evacuation strategy and the impact on urban traffic can be predicted, and the evacuation strategy can be adjusted and optimized according to the traffic flow of each effective path.

[0130] The use of advanced machine learning algorithms and traffic models enables real-time prediction and dynamic adjustment of rail transit emergency passenger flow. This mechanism ensures that in an emergency, the evacuation strategy can quickly adapt to changing passenger flow demand and traffic conditions, improving evacuation efficiency.

[0131] Furthermore, the application of Internet of Things (IoT) technology can collect various sensor data of rail transit systems in real time, improving the real-time and accuracy of predictions. At the same time, social media and network comment analysis can help capture potential emergency information and achieve advance prediction and preparation.

[0132] In this embodiment, multi-source data is collected, wherein the multi-source data includes passenger travel data and mobile phone signaling data; based on historical data analysis and real-time prediction results, the currently monitored passenger evacuation event data is scientifically classified, and corresponding evacuation strategies are formulated. This classification method not only improves the pertinence of emergency response, but also enhances the flexibility and adaptability of the strategy; by integrating multiple data sources, a comprehensive rail transit passenger behavior analysis platform is constructed. This data integration method can provide deeper insights into passenger behavior and provide a scientific basis for the formulation of emergency evacuation strategies;

[0133] Based on the passenger travel data, the probability of each valid path in the OD path is calculated, wherein the OD path is the path between any departure station O and any target station D of the subway; based on the mobile phone signaling data, the number of permanent residents and the number of jobs in the mobile phone signaling grid are calculated; based on the multi-source data, the number of permanent residents and the number of jobs in the mobile phone signaling grid and the probability of each valid path in the OD path, a passenger evacuation strategy is obtained through a traffic model, so as to evacuate passengers based on the passenger evacuation strategy, and adopt advanced machine learning algorithms and traffic models to realize real-time prediction and dynamic adjustment of rail transit emergency passenger flow. This mechanism can ensure that in an emergency, the evacuation strategy can quickly adapt to the changing passenger flow demand and traffic conditions, and improve the evacuation efficiency. Through this embodiment, the technical problems existing in the prior art in terms of responding to emergencies, the flexibility of passenger evacuation strategies and the accuracy of emergency strategies are solved.

[0134] Optionally, in one embodiment, referring to Figure 3 , Figure 3 FIG. 2 is a flow chart of another embodiment of the passenger evacuation method of the present invention. Figure 3 As shown, before step S10, it includes:

[0135] Step S001, obtaining currently monitored passenger evacuation event data and historical passenger evacuation event data;

[0136] In this embodiment, comprehensive data collection of historical emergencies is involved. An accident is taken as a data point. Then, the currently monitored passenger evacuation event data is obtained by analyzing and extracting the event, and the indicator dimensions of the data point are identified. The indicator dimensions can be quantitative data or accident level data. Possible indicator dimensions include the following:

[0137] Event Type: Describes the nature of the event, such as natural disaster (flood, earthquake), accident (train collision, derailment), panic activity, fire, technical failure, etc.

[0138] Event location: The specific location where the event occurred, which can be a rail transit station, line, vehicle, etc.

[0139] Event time: The specific time when the event occurred, including date and time, helps analyze the temporal pattern of events.

[0140] Number of passengers affected: The number of passengers affected by the incident, which helps assess evacuation needs and resource allocation.

[0141] Service interruption time: the duration of rail transit service interruption caused by an incident.

[0142] Emergency response time: the time from the occurrence of an incident to the initiation of emergency response.

[0143] Evacuation efficiency: the efficiency of the evacuation operation, such as the time required to complete the evacuation, the effectiveness of the evacuation route, etc.

[0144] Resource usage: Resources used in emergency response, including personnel, equipment, supplies, etc.

[0145] Event handling results: the final handling results of the event, such as the time to resume normal operations, the results of the event investigation, etc.

[0146] Passenger satisfaction: Passenger satisfaction with emergency response measures can be obtained through questionnaires and other means.

[0147] Economic losses: direct and indirect economic losses caused to rail transit operators by the incident.

[0148] Social impact: The impact of the event on social order and public sentiment, which may include media coverage, social media reactions, etc.

[0149] Environmental impact: the potential impact of an event on the environment, such as pollution, ecological damage, etc.

[0150] Step S002, classifying each historical passenger evacuation event data into accident levels by using a clustering algorithm, and obtaining a corresponding relationship between each historical passenger evacuation event data and the accident level;

[0151] In this embodiment, in the stage of formulating the accident level classification standard, a series of scientific standards need to be set according to each historical passenger evacuation event data. These standards will serve as the basis for the accident level classification. Using these standards, historical events are systematically classified to form different accident levels, and corresponding emergency response strategies are formulated for events of each accident level. In this stage, the K-means clustering algorithm is used to classify the accident levels. The K-means clustering algorithm ensures that the events in the same cluster have a high degree of similarity in characteristics by assigning event data points to different clusters, while the clusters maintain significant differences. The application of this algorithm not only improves the accuracy of event classification, but also provides solid data support for the formulation of targeted emergency strategies. Through this data-driven method, the present invention can ensure that the rail transit system can quickly and accurately classify accidents when facing various emergencies, and take the most appropriate countermeasures.

[0152] The application method of K-means clustering algorithm in accident classification usually involves the following steps, combined with relevant formulas:

[0153] 1) Determine the number of clusters (K): Before applying the K-means algorithm, you need to determine how many clusters (i.e., accident levels) you want to divide your data into. This can be estimated using a variety of methods, such as the Elbow Method or the Silhouette Coefficient.

[0154] 2) Initialize the cluster centers (Centroids): Randomly select K data points as the initial cluster centers. These centroids represent the center of each cluster.

[0155] 3) Assign data points to the nearest cluster center: For each data point, calculate its distance to each cluster center and assign it to the cluster with the nearest cluster center. Distance calculation formula (taking Euclidean distance as an example):

[0156]

[0157] in, is the ith data point, is the i-th cluster center, is the number of features, is the distance between the data point and the cluster center.

[0158] 4) Update the cluster center: For each cluster, calculate the average value of all data points in the cluster and use this average value as the new cluster center . Cluster center update formula:

[0159]

[0160] Where C is the set of all data points in the cluster, is the number of data points in the cluster and x is the data point.

[0161] 5) Iteration process: Repeat steps 3 and 4 until the cluster center no longer changes significantly or the predetermined number of iterations is reached.

[0162] 6) Evaluate clustering results: Use silhouette coefficient or other evaluation indicators to evaluate the quality of clustering.

[0163] 7) Accident classification: According to the clustering results, the accidents are classified into different accident levels, and the corresponding relationship between each historical passenger evacuation event data and the accident level is obtained. For example, if K=4, the accidents can be divided into four accident levels from level 1 to level 4.

[0164] In practical applications, the K-means algorithm may need to be run multiple times to select the best clustering result because the algorithm is sensitive to the choice of initial cluster centers. In addition, for non-spherical clusters or clusters with large size differences, K-means may not be the best choice, and other clustering algorithms such as DBSCAN or spectral clustering may need to be considered.

[0165] Furthermore, for the accident classification module, fuzzy logic or support vector machine (SVM) can be used as an alternative to handle the uncertainty and ambiguity of data. At the same time, the expert system can combine expert knowledge and experience to provide a scientific basis for the classification of incident accident levels.

[0166] Level 1 evacuation (local evacuation):

[0167] For incidents with local impact, evacuation mainly relies on internal station measures.

[0168] Analyze passenger flow trends around the station and optimize evacuation instructions and routes inside the station.

[0169] Level 2 evacuation (area evacuation):

[0170] When the impact of an incident extends to a certain area, the operation of the affected lines needs to be adjusted.

[0171] Based on the rail network passenger flow data, predict the passengers' possible destinations, coordinate with adjacent stations and lines, and ensure smooth evacuation.

[0172] Level 3 evacuation (multi-mode evacuation):

[0173] In larger incidents, it may be necessary to suspend some or all rail services.

[0174] Based on the city's passenger travel characteristics, plan multi-modal transportation evacuation plans, such as buses, taxis, etc., to ensure that passengers can reach their intended destinations safely.

[0175] Level 4 evacuation (full evacuation):

[0176] In extreme cases, rail transit systems may need to suspend service entirely.

[0177] Utilize big data analysis to determine the main evacuation directions and destinations of passengers, and coordinate the city's transportation resources, including ground public transportation, private transportation, and other emergency transportation, to achieve rapid and effective evacuation.

[0178] Based on historical data analysis and real-time prediction results, emergency situations are scientifically classified and corresponding evacuation strategies are formulated, which not only improves the targeted nature of emergency response, but also enhances the flexibility and adaptability of the strategy.

[0179] Step S003: Based on the corresponding relationship, determine the accident level corresponding to the currently monitored passenger evacuation event data.

[0180] In this embodiment, based on the correspondence between each type of historical passenger evacuation event data and the accident level, the accident level corresponding to the currently monitored passenger evacuation event data can be determined.

[0181] In a second aspect, an embodiment of the present invention further provides a passenger evacuation device.

[0182] In one embodiment, referring to Figure 4 , Figure 4 Schematic diagram of the functional modules of an embodiment of the passenger evacuation device of the present invention. Figure 4 As shown, the passenger evacuation device includes:

[0183] The data collection module 10 is configured to collect multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data;

[0184] A first calculation module 20 is configured to calculate the probability of each valid path in the OD path of the rail transit network based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway;

[0185] The second calculation module 30 is configured to calculate the number of permanent residents and the number of jobs in the mobile phone signaling grid based on the mobile phone signaling data;

[0186] The strategy determination module 40 is configured to obtain a passenger flow evacuation strategy through a traffic model based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path.

[0187] Optionally, in one embodiment, the passenger evacuation device further includes an accident level classification module configured to:

[0188] Obtain the currently monitored passenger evacuation event data and historical passenger evacuation event data;

[0189] The accident level of each historical passenger evacuation event data is divided by clustering algorithm, and the corresponding relationship between each historical passenger evacuation event data and the accident level is obtained;

[0190] Based on the corresponding relationship, the accident level corresponding to the currently monitored passenger evacuation event data is determined.

[0191] Optionally, in one embodiment, the first calculation module 20 is configured to:

[0192] Selecting any path in the OD path of the rail transit network as a target path, preprocessing the passenger travel data corresponding to the target path, and obtaining the preprocessed passenger travel data;

[0193] Substitute the preprocessed passenger travel data into the first preset formula to calculate the path cost of the target path. The first preset formula is as follows:

[0194]

[0195] In the formula, represents the path cost of the target path, m represents the number of inter-station paths between two adjacent stations in the OD path, represents the i-th inter-station path in the OD path, represents the running time of the train on the i-th station path in the OD path, n represents the number of transfer station paths in the OD path, represents the transfer factor required by the passenger for the jth transfer path within the OD path, It represents the transfer time required by the passenger to take the jth transfer path within the OD path;

[0196] Select any path from the OD paths of the rail transit network that have not been selected as the target path, return to the step of preprocessing the passenger travel data corresponding to the target path to obtain the preprocessed passenger travel data, until all paths in the OD paths of the rail transit network have been selected, and obtain the path cost of each path in the OD paths of the rail transit network;

[0197] Determining a minimum path cost from the path costs of each of the paths;

[0198] Substitute the minimum path cost into the second preset formula to calculate the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. The second preset formula is as follows:

[0199]

[0200] In the formula, It represents the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. It represents the minimum path cost within the OD path of the rail transit network, a and b are constants, and min() is the minimum value function;

[0201] From the path costs of each path, a path cost that is less than or equal to the maximum path cost threshold is selected as a target path cost, and a path corresponding to the target path cost is taken as a valid path;

[0202] Based on the path costs of the effective paths, the probability of each effective path in the OD path of the rail transit network is calculated.

[0203] Optionally, in one embodiment, the passenger travel data includes: the number of inter-station paths between two adjacent stations in the OD path, the number of transfer station paths in the OD path, the attractiveness correction factor of the i-th inter-station path in the OD path, the running time of the train on the i-th inter-station path in the OD path, the transfer factor required by the passenger for the j-th transfer path in the OD path, and the transfer time required by the passenger for the j-th transfer path in the OD path;

[0204] The preprocessing includes: data filtering processing, formatting processing and normalization processing.

[0205] Optionally, in one embodiment, the first calculation module 20 is configured to:

[0206] For the probability of the kth valid path, the path cost of the kth valid path in the OD path and the minimum path cost in the OD path are substituted into the third preset formula to calculate the probability of the kth valid path in the OD path of the rail transit network. The third preset formula is as follows:

[0207]

[0208] In the formula, represents the probability of the kth valid path in the OD path of the rail transit network, represents the corrected path cost of the kth effective path in the OD path of the rail transit network, , represents the normal parameter, , Represents the path cost of the kth valid path in the OD path of the rail transit network Minimum path cost within the OD path of the rail transit network The ratio of K represents the number of valid paths within the OD path of the rail transit network.

[0209] Optionally, in one embodiment, the second calculation module 30 is configured to:

[0210] Perform spatial data fusion and sampling expansion operations on mobile phone signaling data and permanent population numbers to obtain the permanent population numbers within the mobile phone signaling grid;

[0211] The mobile phone signaling data and the number of positions are spatially fused and expanded to obtain the number of positions in the mobile phone signaling grid.

[0212] Optionally, in one embodiment, the traffic model includes: a travel generation model, a travel distribution model, a traffic mode division model and a traffic assignment model.

[0213] Optionally, in one embodiment, the policy determination module 40 is configured to:

[0214] Based on the number of permanent residents and jobs in the mobile phone signaling grid, the total number of travel population in cell w and the number of travel jobs in cell w are calculated through the travel generation model. Total population entering

[0215] Based on the total number of people traveling in community w and the community The total population entering the community is calculated through the travel distribution model. Number of trips;

[0216] Based on the multi-source data and from cell w to cell The number of trips from community w to community w is calculated by dividing the model by traffic mode. Transportation of passenger traffic;

[0217] Based on the cell w to cell Transportation The passenger flow and the probability of each valid path in the OD path are used to obtain the traffic flow of each valid path through the traffic assignment model;

[0218] Optimize passenger evacuation strategy based on traffic flow of each effective path.

[0219] Optionally, in one embodiment, the policy determination module 40 is configured to:

[0220] The travel generation model is used to obtain the travel rate of the qth travel purpose and the number of permanent residents with the qth travel purpose in the permanent population in the mobile phone signaling grid where the cell w is located;

[0221] Substitute the travel rate of the qth travel purpose and the number of permanent residents with the qth travel purpose in the number of permanent residents in the mobile phone signaling grid where the cell w is located into the fourth preset formula to calculate the total number of travel population in the cell w;

[0222] Obtaining communities through travel generation models The attraction rate of the qth travel purpose among the number of positions in the mobile signaling grid and the number of positions provided by the qth travel purpose;

[0223] The community The attraction rate of the qth type of travel purpose and the number of jobs provided by the qth type of travel purpose in the number of jobs in the mobile phone signaling grid are substituted into the fifth preset formula to calculate the cell Total population entering

[0224] Among them, the fourth preset formula is as follows:

[0225]

[0226] Where: is the total number of people traveling in community w; is the travel rate of the qth travel purpose among the permanent population in the mobile phone signaling grid where cell w is located; is the number of permanent residents with the qth type of travel purpose among the permanent residents in the mobile phone signaling grid where cell w is located;

[0227] The fifth preset formula is as follows:

[0228]

[0229] Where: For the community Total population entering For the community The attraction rate of the qth type of travel purpose among the number of posts in the mobile signaling grid; For the community The number of positions provided by the qth travel purpose among the number of positions within the mobile phone signaling grid.

[0230] Optionally, in one embodiment, the policy determination module 40 is configured to:

[0231] The total number of people traveling in community w and the The total population entering the community is substituted into the travel distribution model to calculate the distance from community w to community The number of trips, the travel distribution model is as follows:

[0232]

[0233] In the formula, From cell w to cell The number of trips, is the total number of people traveling in community w, For the community The total population entering is the impedance function, , From cell w to cell resistance, , and is a constant, is the total number of cells.

[0234] Optionally, in one embodiment, the policy determination module 40 is configured to:

[0235] Based on the multi-source data, the traffic mode classification model is used to calculate the distance from cell w to cell w. Transportation probability;

[0236] Using from cell w to cell Transportation The probability of cell w to cell The number of trips from cell w to cell Transportation of travel passenger flow.

[0237] Among them, the functional implementation of each module in the above-mentioned passenger evacuation device corresponds to the various steps in the above-mentioned passenger evacuation method embodiment, and its functions and implementation processes will not be repeated here one by one.

[0238] In a third aspect, an embodiment of the present invention further provides an electronic device, whose structure is as follows: Figure 5 As shown, it includes: a memory and a processor, wherein the processor is used to read and execute the computer program stored in the memory to implement the above-mentioned passenger evacuation method.

[0239] In a fourth aspect, an embodiment of the present invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and when the computer executable instructions are executed, the aforementioned passenger evacuation method is implemented.

[0240] In a fifth aspect, an embodiment of the present invention provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the various processes of the above-mentioned passenger evacuation method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0241] Finally, it should be noted that: in some processes described in the embodiments of the present invention, multiple operations or steps appearing in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present invention or may be executed in parallel, and the sequence number of the operation is only used to distinguish between different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.

[0242] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A passenger evacuation method, characterized in that: The method comprises: Collecting multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data; The probability of each valid path in the OD path of the rail transit network is calculated based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway; The calculating, based on the passenger travel data, the probability of each valid path in the OD path of the rail transit network includes: Selecting any path in the OD path of the rail transit network as a target path, preprocessing the passenger travel data corresponding to the target path, and obtaining the preprocessed passenger travel data; Substitute the preprocessed passenger travel data into the first preset formula to calculate the path cost of the target path. The first preset formula is as follows: In the formula, represents the path cost of the target path, m represents the number of inter-station paths between two adjacent stations in the OD path, represents the i-th inter-station path in the OD path, represents the running time of the train on the i-th station path in the OD path, n represents the number of transfer station paths in the OD path, represents the transfer factor required by the passenger for the jth transfer path within the OD path, It represents the transfer time required by the passenger to take the jth transfer path within the OD path; Select any path from the OD paths of the rail transit network that have not been selected as the target path, return to the step of preprocessing the passenger travel data corresponding to the target path to obtain the preprocessed passenger travel data, until all paths in the OD paths of the rail transit network have been selected, and obtain the path cost of each path in the OD paths of the rail transit network; Determining a minimum path cost from the path costs of each of the paths; Substitute the minimum path cost into the second preset formula to calculate the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. The second preset formula is as follows: In the formula, It represents the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. It represents the minimum path cost within the OD path of the rail transit network, a and b are constants, and min() is the minimum value function; From the path costs of each path, a path cost that is less than or equal to the maximum path cost threshold is selected as a target path cost, and a path corresponding to the target path cost is taken as a valid path; Based on the path cost of the effective path, the probability of each effective path in the OD path of the rail transit network is calculated; Calculate the number of permanent residents and the number of jobs within the mobile phone signaling grid based on the mobile phone signaling data; Based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, a passenger flow evacuation strategy is obtained through a traffic model, wherein the traffic model includes: a travel generation model, a travel distribution model, a traffic mode division model, and a traffic allocation model; The passenger flow evacuation strategy is obtained through a traffic model based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, including: Based on the number of permanent residents and jobs in the mobile phone signaling grid, the total number of travel population in cell w and the number of travel jobs in cell w are calculated through the travel generation model. Total population entering Based on the total number of people traveling in community w and the community The total population entering the community is calculated through the travel distribution model. Number of trips; Based on the multi-source data and from cell w to cell The number of trips from community w to community w is calculated by dividing the model by traffic mode. Transportation of passenger traffic; Based on the cell w to cell Transportation The passenger flow and the probability of each valid path in the OD path are used to obtain the traffic flow of each valid path through the traffic assignment model; Optimize passenger evacuation strategy based on traffic flow of each effective path.

2. The passenger evacuation method according to claim 1, characterized in that: Before collecting multi-source data, determine the accident level corresponding to the current passenger evacuation event data, including: Obtain the currently monitored passenger evacuation event data and historical passenger evacuation event data; The accident level of each historical passenger evacuation event data is divided by clustering algorithm, and the corresponding relationship between each historical passenger evacuation event data and the accident level is obtained; Based on the corresponding relationship, the accident level corresponding to the currently monitored passenger evacuation event data is determined.

3. The passenger evacuation method according to claim 1, characterized in that: The passenger travel data includes: the number of inter-station paths between two adjacent stations in the OD path, the number of transfer station paths in the OD path, the attractiveness correction factor of the i-th inter-station path in the OD path, the running time of the train on the i-th inter-station path in the OD path, the transfer factor required by the passenger for the j-th transfer path in the OD path, and the transfer time required by the passenger for the j-th transfer path in the OD path; The preprocessing includes: data filtering processing, formatting processing and normalization processing.

4. The passenger evacuation method according to claim 1, characterized in that: The step of calculating the probability of each valid path in the OD path of the rail transit network based on the path cost of the valid path includes: For the probability of the kth valid path, the path cost of the kth valid path in the OD path and the minimum path cost in the OD path are substituted into the third preset formula to calculate the probability of the kth valid path in the OD path of the rail transit network. The third preset formula is as follows: In the formula, represents the probability of the kth valid path in the OD path of the rail transit network, represents the corrected path cost of the kth effective path in the OD path of the rail transit network, , represents the normal parameter, , Represents the path cost of the kth valid path in the OD path of the rail transit network Minimum path cost within the OD path of the rail transit network The ratio of K represents the number of valid paths within the OD path of the rail transit network.

5. The passenger evacuation method according to claim 1, characterized in that: The step of calculating the number of permanent residents and the number of jobs in the mobile phone signaling grid based on the mobile phone signaling data comprises: Perform spatial data fusion and sampling expansion operations on mobile phone signaling data and permanent population numbers to obtain the permanent population numbers within the mobile phone signaling grid; The mobile phone signaling data and the number of positions are spatially fused and expanded to obtain the number of positions in the mobile phone signaling grid.

6. The passenger evacuation method according to claim 1, characterized in that: Based on the number of permanent residents and the number of jobs in the mobile phone signaling grid, the total number of travel population in cell w and the number of travel jobs in cell w are calculated through the travel generation model. The steps to enter the total population include: The travel generation model is used to obtain the travel rate of the qth travel purpose and the number of permanent residents with the qth travel purpose in the permanent population in the mobile phone signaling grid where the cell w is located; Substitute the travel rate of the qth travel purpose and the number of permanent residents with the qth travel purpose in the number of permanent residents in the mobile phone signaling grid where the cell w is located into the fourth preset formula to calculate the total number of travel population in the cell w; Obtaining communities through travel generation models The attraction rate of the qth travel purpose among the number of positions within the mobile signaling grid and the number of positions provided by the qth travel purpose; The community The attraction rate of the qth type of travel purpose and the number of jobs provided by the qth type of travel purpose in the number of jobs in the mobile phone signaling grid are substituted into the fifth preset formula to calculate the cell Total population entering Among them, the fourth preset formula is as follows: Where: is the total number of people traveling in community w; is the travel rate of the qth travel purpose among the permanent population in the mobile phone signaling grid where cell w is located; is the number of permanent residents with the qth type of travel purpose among the permanent residents in the mobile phone signaling grid where cell w is located; The fifth preset formula is as follows: Where: For the community Total population entering For the community The attraction rate of the qth type of travel purpose among the number of posts in the mobile signaling grid; For the community The number of positions provided by the qth travel purpose among the number of positions within the mobile phone signaling grid.

7. The passenger evacuation method according to claim 1, characterized in that: The total number of people traveling based on the community w and the community The total population entering the community is calculated through the travel distribution model. The steps to calculate the number of trips include: The total number of people traveling in community w and the The total population entering the community is substituted into the travel distribution model to calculate the distance from community w to community The number of trips, the travel distribution model is as follows: In the formula, From cell w to cell The number of trips, is the total number of people traveling in community w, For the community The total population entering is the impedance function, , From cell w to cell resistance, , and is a constant, is the total number of cells.

8. The passenger evacuation method according to claim 1, characterized in that: Based on the multi-source data and from cell w to cell The number of trips from community w to community w is calculated by dividing the model by traffic mode. Transportation The steps to measure the travel passenger flow include: Based on the multi-source data, the traffic mode classification model is used to calculate the distance from cell w to cell w. Transportation The probability of Using from cell w to cell Transportation The probability of cell w to cell The number of trips from cell w to cell Transportation of travel passenger flow.

9. A passenger evacuation device, characterized in that: The device comprises: A data collection module is configured to collect multi-source data, wherein the multi-source data includes passenger travel data and mobile phone signaling data; A first calculation module is configured to calculate the probability of each valid path in the OD path of the rail transit network based on the passenger travel data, wherein the OD path is a path between any departure station O and any destination station D of the subway; The first computing module is specifically configured to: Selecting any path in the OD path of the rail transit network as a target path, preprocessing the passenger travel data corresponding to the target path, and obtaining the preprocessed passenger travel data; Substitute the preprocessed passenger travel data into the first preset formula to calculate the path cost of the target path. The first preset formula is as follows: In the formula, represents the path cost of the target path, m represents the number of inter-station paths between two adjacent stations in the OD path, represents the i-th inter-station path in the OD path, represents the running time of the train on the i-th station path in the OD path, n represents the number of transfer station paths in the OD path, represents the transfer factor required by the passenger for the jth transfer path within the OD path, It represents the transfer time required by the passenger to take the jth transfer path within the OD path; Select any path from the OD paths of the rail transit network that have not been selected as the target path, return to the step of preprocessing the passenger travel data corresponding to the target path to obtain the preprocessed passenger travel data, until all paths in the OD paths of the rail transit network have been selected, and obtain the path cost of each path in the OD paths of the rail transit network; Determining a minimum path cost from the path costs of each of the paths; Substitute the minimum path cost into the second preset formula to calculate the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. The second preset formula is as follows: In the formula, It represents the maximum path cost threshold acceptable to passengers in the OD path of the rail transit network. It represents the minimum path cost within the OD path of the rail transit network, a and b are constants, and min() is the minimum value function; From the path costs of each path, a path cost that is less than or equal to the maximum path cost threshold is selected as a target path cost, and a path corresponding to the target path cost is taken as a valid path; Based on the path cost of the effective path, the probability of each effective path in the OD path of the rail transit network is calculated; A second calculation module is configured to calculate the number of permanent residents and the number of jobs in the mobile phone signaling grid based on the mobile phone signaling data; A strategy determination module is configured to obtain a passenger flow evacuation strategy through a traffic model based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, wherein the traffic model includes: a travel generation model, a travel distribution model, a traffic mode division model, and a traffic allocation model; The policy determination module is specifically configured to: The passenger flow evacuation strategy is obtained through a traffic model based on the multi-source data, the number of permanent residents in the mobile phone signaling grid, the number of jobs, and the probability of each valid path in the OD path, including: Based on the number of permanent residents and jobs in the mobile phone signaling grid, the total number of travel population in cell w and the number of travel jobs in cell w are calculated through the travel generation model. Total population entering Based on the total number of people traveling in community w and the community The total population entering the community is calculated through the travel distribution model. Number of trips; Based on the multi-source data and from cell w to cell The number of trips from community w to community w is calculated by dividing the model by traffic mode. Transportation of passenger traffic; Based on the cell w to cell Transportation The passenger flow and the probability of each valid path in the OD path are used to obtain the traffic flow of each valid path through the traffic assignment model; Optimize passenger evacuation strategy based on traffic flow of each effective path.

10. An electronic device, characterized in that: include: Memory and processor; The processor is used to read and execute the computer program stored in the memory to implement the steps of the passenger evacuation method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed, the steps of the passenger evacuation method as described in any one of claims 1-8 are implemented.

12. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the passenger evacuation method as described in any one of claims 1-8 are implemented.