Urban rail transit passenger flow guidance method and safety management system based on graph theory

Abstract

The invention relates to the technical field of rail transit, in particular to an urban rail transit passenger flow guidance method and safety management system based on a graph theory, and the method comprises the steps: building a network topological graph based on a spatial topological relation in a rail transit station, and obtaining a passenger flow guidance result through a monitoring video image of each road section in the rail transit station in real time; obtaining the passenger flow volume of each road section at different moments; calculating a passage selectable value, a passage state value, a passage efficiency value and an induction weight of each road section at the current moment; and a weighted network topological graph at the current moment is obtained, a dynamic path planning algorithm is combined, an optimal induction path is obtained, and passengers in the rail transit station are induced to reasonably select a passing path. The optimal induction path is updated in real time, accurate planning of the induction path is improved, dense distribution of passengers in the station is avoided, potential safety hazards are effectively prevented and relieved, and the passing safety of the passengers is improved.

Description

Technical Field

[0001] This application relates to the technical field of rail transit, and particularly to an urban rail transit passenger flow induction method and a safety management system based on graph theory. Background Art

[0002] With the gradual expansion of the scale of rail transit, large passenger flows have become the norm in rail transit operations. Large passenger flows in rail transit stations not only cause congestion within the stations, affecting the passing efficiency of passengers, but also may lead to stampede incidents and pose great potential safety hazards if effective guiding and evacuation measures are not taken.

[0003] For urban rail transits such as subways and light rails, the current pedestrian guiding information provided to passengers within stations is mainly static guiding signs, which do not consider the real-time dynamic impact of the periodic arrival of trains on the passenger flow, easily leading to deviations in the judgment of the passenger flow, affecting the rationality and accuracy of route planning, resulting in poor induction effects, causing dense distribution of passengers within the stations and prone to potential safety hazards. Summary of the Invention

[0004] In order to solve the above technical problems, an urban rail transit passenger flow induction method and a safety management system based on graph theory are provided to solve the existing problems.

[0005] The solution of this application to solve the technical problems is to provide an urban rail transit passenger flow induction method and a safety management system based on graph theory, including the following steps: In the first aspect, an embodiment of this application provides an urban rail transit passenger flow induction method based on graph theory, and this method includes the following steps: Based on the spatial topological relationship within the rail transit station, construct a network topological graph, and obtain the passenger flow of each section at different times in real time through the monitoring video images of each section within the rail transit station; Analyze the periodic changes in the passenger flow of each section in the historical period, and combine with the predicted trend of the passenger flow of each section at the current moment to calculate the passing option value of each section at the current moment; Denote the current moment and multiple moments before it as a local period, and calculate the passing state value of each section at the current moment through the change trend and its discreteness of the passenger flow of each section within the local period; Based on the passing option value and the passing state value, determine the passing efficiency value of each section at the current moment; Based on the distance characteristics of each section, combine with the passing efficiency value to determine the induction weight of each section at the current moment; Based on the induction weight, obtain the weighted network topological graph at the current moment, and combine with the dynamic path planning algorithm to obtain the optimal induction path, guiding passengers within the rail transit station to reasonably select passing paths.

[0006] Preferably, target detection is performed on the monitored video images to obtain the passenger flow of each road section at different times.

[0007] Preferably, the calculation of the passage option value of each road section at the current time includes: Perform trend decomposition on the passenger flow of each road section at all times in the historical period, and calculate the periodic intensity of each road section; Perform passenger flow prediction through the change trend of the passenger flow of each road section in the historical period, and determine the predicted passenger flow of each road section at the current time; Take the ratio of the periodic intensity to the predicted passenger flow as the passage option value of each road section at the current time.

[0008] Preferably, the determination process of the predicted passenger flow is as follows: Based on the passenger flow of each road section at all times in the historical period, use a prediction algorithm to predict the passenger flow of each road section at multiple times after the current time, and take the average value of the predicted passenger flows of each road section at multiple times after the current time as the predicted passenger flow of each road section at the current time.

[0009] Preferably, the calculation of the passage state value of each road section at the current time includes: Calculate the dispersion degree of the passenger flow of each road section at all times within the local time period at the current time; Calculate the change rate of the passenger flow between each moment and the previous moment of each road section within the local time period; take the average value of all the change rates of each road section within the local time period as the average change rate of each road section at the current time; The passage state value is the normalized result of the product of the average change rate and the dispersion degree.

[0010] Preferably, the passage efficiency value is the ratio of the passage option value to the passage state value.

[0011] Preferably, the determination of the induction weight of each road section at the current time includes: Calculate the distance length of each road section; perform negative mapping processing on the distance length of each road section; Through a preset first weight and a preset second weight, perform weighted summation on the passage efficiency value and the result of the negative mapping processing to obtain the induction weight of each road section at the current time, where the sum of the preset first weight and the preset second weight is 1, and the preset first weight is greater than the preset second weight.

[0012] Preferably, the method for obtaining the weighted network topology map at the current time is: Each road segment corresponds to each edge in the network topology graph. Taking the induced weight as the weight of each edge in the network topology graph, a weighted network topology graph at the current moment is obtained.

[0013] Preferably, obtaining the optimal induced path includes: Through the GPS positioning system, obtain the starting point and the ending point of the passenger at the current moment; Based on the starting point and the ending point, and the weighted network topology graph, calculate the optimal induced path through the dynamic path planning algorithm.

[0014] In a second aspect, the embodiment of the present application also provides an urban rail transit safety management system based on graph theory. The system includes: A road segment monitoring module, configured to construct a network topology graph based on the spatial topology relationship within the rail transit station, and obtain the passenger flow volume of each road segment at different moments in real time through the monitoring video images of each road segment within the rail transit station; A path analysis module analyzes the periodic changes in the passenger flow volume of each road segment in the historical period, combines with the predicted trend of the passenger flow volume of each road segment at the current moment, and calculates the passing option value of each road segment at the current moment; Denote the current moment and multiple moments before it as a local time period. Calculate the passing state value of each road segment at the current moment through the change trend and the discrete situation of the passenger flow volume of each road segment within the local time period; Based on the passing option value and the passing state value, determine the passing efficiency value of each road segment at the current moment; Based on the distance characteristics of each road segment, combine with the passing efficiency value, determine the induced weight of each road segment at the current moment; Based on the induced weight, obtain a weighted network topology graph at the current moment, and combine with the dynamic path planning algorithm to obtain the optimal induced path; A passenger flow induction module is configured to publish induction information to passengers through the optimal induced path to induce passengers within the rail transit station to reasonably select a passing path.

[0015] The present application has at least the following beneficial effects: According to the spatial topological relationship within the station, this application constructs a network topology graph. By analyzing the periodic variation patterns of the passenger flow on the road sections corresponding to each edge in the network topology graph and the analysis results of predicting the passenger flow on each road section, it calculates the passing option values for each road section at the current moment. The beneficial effects are as follows: it considers the distribution trend of the passenger flow on each road section in a future period of time and combines the periodic variation degree of the passenger flow in the historical period, so as to more accurately reflect the accuracy of the passenger flow prediction and evaluate the passing efficiency of each road section in the subsequent time period; secondly, by the fluctuation situation of the passenger flow within the local time period at the current moment and the change rate of the passenger flow, it calculates the passing state values for each road section at the current moment. The beneficial effects are as follows: it considers the speed of increase or decrease of the passenger flow on each road section to illustrate the moving speed of the passengers within that road section, and further reflects the smoothness of passing on that road section; it determines the passing efficiency values for each road section at the current moment and obtains the induction weights for each road section at the current moment. The beneficial effects are as follows: it considers the passing efficiency of the passengers on each road section and the distance length of the road section to evaluate the possibility of inducing passengers to choose that road section for passing; it constructs a weighted network topology graph, calculates the optimal induction path through a dynamic path planning algorithm, and publishes induction information to passengers to induce passengers to reasonably choose the passing path. The beneficial effects are as follows: it considers the periodic impact of the periodic train arrivals on the passenger flow, can update the optimal induction path in real time by analyzing the changes in the passenger flow in real time, reduce the evaluation deviation of the passenger flow, improve the accuracy of planning the induction path, avoid the dense distribution of passengers in the station, effectively prevent and alleviate the occurrence of potential safety hazards, and improve the passing safety of passengers. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The following further elaborates in detail on the urban rail transit passenger flow induction method based on graph theory of this application with reference to the accompanying drawings.

[0017] Figure 1 It is the flowchart of the steps of the urban rail transit passenger flow induction method based on graph theory provided by the embodiment of this application; Figure 2 It is the flowchart of the steps of the method for obtaining the passing option values for each road section at the current moment provided by the embodiment of this application; Figure 3 It is the flowchart of the steps of the method for obtaining the passing state values for each road section at the current moment provided by the embodiment of this application; Figure 4 It is the block diagram of the urban rail transit safety management system based on graph theory provided by the embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To make the objectives, technical solutions, and advantages of this application more clear and understandable, the following further elaborates on the method for inducing passenger flow and the safety management system based on graph theory proposed in this application in combination with the accompanying drawings and implementation examples. It should be understood that the specific implementation examples described herein are only used to explain this application and are not used to limit this application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs.

[0020] Please refer to Figure 1 , which shows the flowchart of the steps of the method for inducing passenger flow based on graph theory provided in an embodiment of this application. The method includes the following steps: Step 1, based on the spatial topological relationship within the rail transit station, construct a network topology graph, and obtain the passenger flow of each section at different times in real-time through the monitoring video images of each section within the rail transit station.

[0021] Graph theory is a mathematical branch specifically for studying graphs, providing a language for describing networks and a platform for research. Through the study of graphs, relevant topological properties of specific actual networks can be obtained. Therefore, the spatial topological relationship inside the subway station in urban rail transit is represented by the method of graph theory.

[0022] Thus, through the spatial topological relationship within the subway station, each entrance and exit, security check gate, staircase or elevator, and waiting area inside the subway station are abstracted as nodes, and the sections that can be passed between different nodes are abstracted as edges, thereby forming a network topology graph.

[0023] Secondly, each section is monitored in real-time through the monitoring inside the subway station. Using the object detection algorithm, passengers in the monitoring video of each section are detected, and the passenger flow of each section is obtained in real-time; In this embodiment, the Faster R-CNN algorithm is used for object detection. Among them, the Faster R-CNN algorithm is a well-known technology and will not be elaborated here. As other implementation manners, implementers can use other methods of existing technologies, such as the YOLOV5 algorithm, etc. This embodiment does not make special restrictions on this; secondly, in this embodiment, the passenger flow of each section is counted every 1 minute.

[0024] It should be noted that each section corresponds to each edge in the network topology graph.

[0025] So far, the passenger flow of each section and the network topology graph are obtained in real-time.

[0026] Step 2: Analyze the periodic changes in the passenger flow of each section during the historical period, and calculate the traffic option value of each section at the current moment in combination with the predicted trend of the passenger flow of each section at the current moment.

[0027] Under normal circumstances, when inducing the passenger flow, it is generally necessary to consider the passenger flow and distance of each section, and try to let passengers pass through the section with less passenger flow, less travel time and the shortest distance. However, the change speed of the passenger flow in the subway station is relatively fast. Especially when the train enters the station, a large number of passengers will enter the train, and a large number of passengers will also get off the train. Since the arrival time of each train is fixed, therefore, it is analyzed that the passenger flow in the subway station will change periodically. The passenger flow of each section is predicted according to the passenger flow change in the historical period, and then the traffic option value of each section is calculated. The step flow chart of the method for obtaining the traffic option value of each section at the current moment provided by the embodiment of the present application is as Figure 2 shown, specifically including: Obtain the passenger flow of each section at each moment during the historical period through the monitoring video images of several historical days; In this embodiment, it is assumed that the operation time of the subway station is from 6:00 to 23:00, then obtain the passenger flow of each section at each moment within 6:00 to 23:00; Therefore, select the monitoring videos of a total of 7 days in a week, and obtain the passenger flow of each section at each moment during the historical period through the target detection algorithm. As other implementation manners, the implementer can set it by himself according to the actual situation.

[0028] Perform trend decomposition on the passenger flow of each section at all moments during the historical period, and calculate the periodic intensity of each section; In this embodiment, the STL (Seasonal and Trend decomposition using Loess) algorithm is used for trend decomposition. Among them, the STL algorithm is a well-known technology and will not be elaborated here; Secondly, through the STL algorithm, the passenger flow of each section at all moments during the historical period is decomposed into a seasonal sequence and a residual sequence. Among them, the seasonal sequence usually reflects the periodic change of the passenger flow. Therefore, the calculation of the periodic intensity is the same as the calculation of the seasonal intensity. The calculation formula of the periodic intensity is: , where is the periodic intensity, is the variance of the residual sequence, is the variance of the seasonal sequence and the residual sequence, is to find the maximum value; among them, the calculation formula of the seasonal intensity is a well-known technology and will not be elaborated here.

[0029] It should be noted that the greater the periodic intensity, the stronger the regularity of the passenger flow change on this section in the subway station.

[0030] Based on the passenger flow of each road section at all times during the historical period, the passenger flow of each road section at multiple times after the current moment is predicted through a prediction algorithm, and the average value of the predicted passenger flow of each road section at multiple times after the current moment is used as the predicted passenger flow of each road section at the current moment; In this embodiment, a long short-term memory network model (LSTM) is used for prediction. Among them, the long short-term memory network model is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of the prior art, for example, BP neural network model, ARIMA model, etc. This embodiment does not make special restrictions on this; secondly, obtain the predicted passenger flow of each road section at all times within 7 minutes after the current moment. As other implementation manners, implementers can set it by themselves according to the actual situation.

[0031] Take the ratio of the periodic intensity to the predicted passenger flow as the traffic selection value of each road section at the current moment; It should be noted that the smaller the predicted passenger flow, the smaller the passenger flow of the road section at subsequent moments, and the larger the obtained traffic selection value, which reflects that the traffic of this road section is smoother at this time, and passengers should be induced to select this road section for passing.

[0032] Thus, the traffic selection value of each road section at the current moment is obtained.

[0033] Step 3, record the current moment and multiple moments before it as a local time period, and calculate the traffic state value of each road section at the current moment according to the change trend and dispersion of the passenger flow of each road section within the local time period.

[0034] Furthermore, if there are continuously passengers entering the subway station but the train still has a long time to arrive, there will be a situation where the passenger flow inside the station only increases and does not decrease for a period of time. Then the passenger flow in different areas inside the station is in a large state. After passenger flow induction, a larger passenger flow appears in a local area, resulting in an increased probability of potential safety hazards.

[0035] Secondly, among the passengers entering the subway station, one type of passenger is for traveling by train. This type of passenger moves relatively slowly in the face of a large passenger flow; another type of passenger commutes between two entrances and exits in the station to reach the other side of the road. This type of passenger can quickly exit from another exit after entering the station, and the moving speed of the passenger is relatively fast. By analyzing the change speed of the passenger flow on different road sections and calculating the traffic state value, the step flowchart of the method for obtaining the traffic state value of each road section at the current moment provided by the embodiment of the present application is as Figure 3 shown, specifically including: Denote the current moment and multiple moments before it as the local time period. In this embodiment, the current moment and the 7 moments before it are denoted as the local time period, where the value 7 is selected according to the departure time interval of trains in urban rail transit. As other implementation manners, the implementer can set it by himself according to the actual situation.

[0036] Calculate the dispersion degree of the passenger flow volume of each road section at all moments within the local time period at the current moment. In this embodiment, the dispersion degree is measured by calculating the standard deviation of the passenger flow volume of each road section at all moments within the local time period at the current moment. As other implementation manners, the implementer can adopt other methods of the prior art, such as variance, mean absolute deviation, etc. This embodiment does not make special restrictions on this.

[0037] Calculate the change rate of the passenger flow volume between two adjacent moments within the local time period for each road section at the current moment; take the mean value of all the change rates of each road section at the current moment within the local time period as the average change rate of each road section at the current moment. It should be noted that the calculation of the change rate is a well-known technology and will not be elaborated here.

[0038] Take the normalized result of the product of the average change rate and the dispersion degree as the traffic state value of each road section at the current moment. In this embodiment, the sigmoid function is used for normalization processing. The sigmoid function is a well-known technology and will not be elaborated here. As other implementation manners, the implementer can adopt other methods of the prior art, such as the tanh function, etc. This embodiment does not make special restrictions on this.

[0039] It should be noted that the larger the average change rate is and the larger its absolute value is, it indicates that the growth rate of the passenger flow volume on this road section is faster at this time, which reflects that the passenger flow volume increases significantly, the passenger flow on this road section is dense, and the traffic is relatively congested; the smaller the average change rate is and the larger its absolute value is, it indicates that the reduction rate of the passenger flow volume on this road section is faster at this time, which reflects that passengers can pass quickly on this road section; the larger the dispersion degree is, it indicates that the fluctuation of the passenger flow volume within the local time period is larger. Therefore, after normalization processing, the smaller the obtained traffic state value is, it indicates that the reduction rate of the passenger flow on this road section is larger, which reflects that the traffic on this road section is smoother, and passengers should be guided to enter the road section with smoother traffic to facilitate their quick access to the waiting area.

[0040] Thus, the traffic state value of each road section at the current moment is obtained.

[0041] Step 4: Based on the passage optional values and the passage status values, determine the passage efficiency value of each section at the current moment; based on the distance characteristics of each section, combine with the passage efficiency value to determine the guidance weight of each section at the current moment; based on the guidance weight, obtain the weighted network topology graph at the current moment, and combine with the dynamic path planning algorithm to obtain the optimal guidance path, so as to guide the passengers in the rail transit station to reasonably select the passage path.

[0042] Further, based on the passage optional values and the passage status values, determine the passage efficiency value, specifically: Take the ratio of the passage optional value to the passage status value as the passage efficiency value of each section at the current moment; It should be noted that the larger the passage efficiency value is, the higher the passage efficiency of this section is, and the more the passenger flow should be induced to select this section for passage.

[0043] Secondly, in order to enable passengers to pass through the sections with less passenger flow and less passage time, based on the passage efficiency value and the distance of each section, calculate the guidance weight of each section, specifically: Calculate the distance length of each section; Perform negative mapping processing on the distance lengths of all sections; In this embodiment, the process of negative mapping is: calculate the sum of the maximum value and the minimum value of the distance lengths of all sections; calculate the difference between the sum and the distance of each section.

[0044] Through the preset first weight and the preset second weight, perform weighted summation on the passage efficiency value and the result of the negative mapping processing to obtain the guidance weight of each section at the current moment, where the sum of the preset first weight and the preset second weight is 1, and the preset first weight is greater than the preset second weight; In this embodiment, the preset first weight takes the value of 0.7, and the preset second weight takes the value of 0.3. As other implementation manners, the implementer can set them according to the actual situation.

[0045] It should be noted that the larger the guidance weight is, the more the passenger flow should be induced to select this section for passage.

[0046] Each section corresponds to each edge in the network topology graph. Take the guidance weight as the weight of each edge in the network topology graph to obtain the weighted network topology graph at the current moment; Through the GPS positioning system, the starting point and the ending point of the passenger at the current moment are obtained, and through the dynamic path planning algorithm, the optimal guiding path is calculated, and guiding information is issued to the passenger, or the recommended path is displayed through the electronic display screen in the station, guiding the passenger to reasonably select the passing path, thereby avoiding the dense distribution of passengers in the station, effectively preventing and alleviating the occurrence of potential safety hazards, and improving the passing safety of passengers.

[0047] In this embodiment, the dynamic path planning algorithm adopts the D*Lite algorithm. Among them, the D*Lite algorithm is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of the existing technology. For example, the Dijkstra algorithm, etc. This embodiment does not make special restrictions on this.

[0048] Based on the same inventive concept as the above method, the embodiment of the present application also provides an urban rail transit safety management system based on graph theory. The system includes: A section monitoring module, which is used to construct a network topology graph based on the spatial topology relationship in the rail transit station, and obtain the passenger flow of each section at different times in real time through the monitoring video images of each section in the rail transit station; A path analysis module analyzes the periodic changes in the passenger flow of each section in the historical period, combines the predicted trend of the passenger flow of each section at the current moment, and calculates the passing optional value of each section at the current moment; The current moment and multiple moments before it are recorded as a local time period. Through the change trend and its discreteness of the passenger flow of each section in the local time period, the passing state value of each section at the current moment is calculated; Based on the passing optional value and the passing state value, the passing efficiency value of each section at the current moment is determined; Based on the distance characteristics of each section, combined with the passing efficiency value, the guiding weight of each section at the current moment is determined; Based on the guiding weight, the weighted network topology graph at the current moment is obtained, and combined with the dynamic path planning algorithm, the optimal guiding path is obtained; A passenger flow guiding module is used to issue guiding information to passengers through the optimal guiding path, guiding the passengers in the rail transit station to reasonably select the passing path.

[0049] Among them, the block diagram of the urban rail transit safety management system based on graph theory provided by the embodiment of the present application is as Figure 4 shown.

[0050] It should be understood that although Figure 1 the steps in the flowchart of Figure 1At least some of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages do not necessarily need to be completed at the same moment, but can be executed at different moments. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least some of the other steps or sub-steps or stages of the other steps.

[0051] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as within the scope described in this specification.

[0052] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation to the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made. Therefore, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application all fall within the protection scope of the technical solution of the present application.

Claims

1. A method for inducing passenger flow in urban rail transit based on graph theory, characterized in that: The method comprises the following steps: Based on the spatial topological relationship in the rail transit station, a network topology map is constructed, and the passenger flow of each section at different times is obtained through real-time monitoring video images of each section in the rail transit station; Analyze the periodic changes of passenger flow of each road section in the historical period, and calculate the optional value of each road section at the current moment based on the predicted trend of passenger flow of each road section at the current moment; The current moment and the previous moments are recorded as a local period, and the traffic state value of each road section at the current moment is calculated through the change trend and discreteness of the passenger flow of each road section in the local period; Determine the traffic efficiency value of each road section at the current moment based on the traffic optional value and the traffic status value; Based on the distance characteristics of each road section and in combination with the traffic efficiency value, determining the induction weight of each road section at the current moment; Based on the induction weights, a weighted network topology diagram at the current moment is obtained, and combined with a dynamic path planning algorithm, an optimal induction path is obtained to induce passengers in the rail transit station to reasonably choose a travel path.

2. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: Target detection is performed on the surveillance video image to obtain the passenger flow of each road section at different times.

3. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: The calculation of the passable optional value of each road section at the current moment includes: The trend of passenger flow of each road section at all times in the historical period is decomposed to calculate the periodic intensity of each road section; Passenger flow forecasting is performed based on the changing trend of passenger flow of each road section in the historical period to determine the predicted passenger flow of each road section at the current moment; The ratio of the periodic intensity to the predicted passenger flow is used as the optional value for traffic on each road section at the current moment.

4. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 3, characterized in that: The process of determining the predicted passenger flow is as follows: Based on the passenger flow of each road section at all times in the historical period, the passenger flow of each road section at multiple times after the current moment is predicted through a prediction algorithm, and the average of the predicted passenger flow of each road section at multiple times after the current moment is taken as the predicted passenger flow of each road section at the current moment.

5. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: The calculation of the traffic status value of each road section at the current moment includes: Calculate the degree of dispersion of passenger flow at all times in the local time period for each road section at the current time; Calculate the change rate of passenger flow of each road section between each moment in the local time period and the previous moment; take the average of all the change rates of each road section in the local time period as the average change rate of each road section at the current moment; The traffic state value is a normalized result of the product of the average change rate and the discrete degree.

6. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: The traffic efficiency value is the ratio of the traffic optional value to the traffic status value.

7. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: Determining the induction weight of each road section at the current moment includes: Calculate the distance length of each road section; perform negative mapping on the distance length of each road section; By presetting the first weight and the second weight, the traffic efficiency value and the result of the negative mapping processing are weighted and summed to obtain the induced weight of each road section at the current moment, wherein the sum of the preset first weight and the preset second weight is 1, and the preset first weight is greater than the preset second weight.

8. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: The method for obtaining the weighted network topology map at the current moment is: Each road section corresponds to each edge in the network topology diagram, and the induced weight is used as the weight of each edge in the network topology diagram to obtain the weighted network topology diagram at the current moment.

9. The urban rail transit passenger flow induction method based on graph theory as claimed in claim 1, characterized in that: The obtaining of the optimal induction path comprises: Obtain the starting and ending points of passengers at the current moment through the GPS positioning system; Based on the starting point and the end point, and the weighted network topology graph, an optimal induced path is calculated by a dynamic path planning algorithm.

10. A graph-theory-based urban rail transit safety management system, applying the graph-theory-based urban rail transit passenger flow induction method in claim 1, characterized in that: The system comprises: The road section monitoring module is used to construct a network topology map based on the spatial topological relationship within the rail transit station, and obtain the passenger flow of each road section at different times through the monitoring video images of each road section in the rail transit station in real time; The path analysis module analyzes the periodic changes in passenger flow of each road section in the historical period, and calculates the traffic option value of each road section at the current moment in combination with the predicted trend of passenger flow of each road section at the current moment; the current moment and multiple moments before it are recorded as local time periods, and the traffic state value of each road section at the current moment is calculated through the change trend and discrete situation of passenger flow of each road section in the local time period; based on the traffic option value and the traffic state value, the traffic efficiency value of each road section at the current moment is determined; based on the distance characteristics of each road section, combined with the traffic efficiency value, the induction weight of each road section at the current moment is determined; based on the induction weight, the weighted network topology diagram at the current moment is obtained, and combined with the dynamic path planning algorithm, the optimal induction path is obtained; The passenger flow induction module is used to issue induction information to passengers through the optimal induction path, inducing passengers in the rail transit station to reasonably choose the passage path.

Images