Rail transit intelligent scheduling method and system

Abstract

The invention discloses a rail transit intelligent scheduling method and system, and relates to the field of rail transit, and the method comprises the steps: obtaining an active passenger flow arrival prediction result of a target venue, dividing a platform space into a plurality of virtual waiting areas according to the active passenger flow arrival prediction result, and configuring a departure condition combination library for each virtual waiting area, obtaining the real-time number of waiting people and waiting time of each virtual waiting area, and calculating the congestion index of the current waiting area according to the real-time number of waiting people and waiting time; a matched target departure condition is selected from a departure condition combination library according to the congestion degree index, when the number of waiting people and the waiting time meet the target departure condition, a departure instruction is triggered, and the operation efficiency score after each departure is calculated; the area with the efficiency score higher than a preset threshold value is set as a rapid departure area, and the area with the efficiency score lower than the preset threshold value is set as a conventional departure area. By implementing the method, the transport capacity resource utilization efficiency of the rail transit can be improved.

Description

Technical Field

[0001] The present application relates to the field of rail transit, and in particular to a rail transit intelligent scheduling method and system. Background Art

[0002] With the acceleration of urbanization and the frequent holding of large-scale events, rail transit stations around the venues are facing the pressure of a sharp increase in passenger flow during events. This kind of sudden high-density passenger flow puts higher requirements on rail transit operation and scheduling, requiring the rail transit system to respond quickly and reasonably allocate transportation resources.

[0003] At present, the rail transit industry mainly adopts a dispatching method that combines fixed timetables and manual experience judgment. The basic operation diagram is formulated based on historical passenger flow data. When there is a large fluctuation in passenger flow, the dispatcher will make manual intervention adjustments based on on-site feedback. Some lines also adopt an automated dispatching mode based on fixed departure intervals, setting different departure interval parameters according to time periods.

[0004] This scheduling method performs stably in daily operations. However, when faced with sudden high-density passenger flow caused by venue activities, due to the lack of refined spatial management and dynamic response mechanism, it is easy to cause people to gather in local areas of the platform, while other areas are relatively idle, resulting in low efficiency in the overall utilization of transportation resources. At the same time, fixed departure strategies are difficult to make differentiated adjustments based on the actual passenger flow characteristics of different areas, resulting in long waiting times or excess capacity in some areas. Summary of the invention

[0005] The present application provides a rail transit intelligent scheduling method and system for improving the utilization efficiency of rail transit capacity resources.

[0006] In a first aspect, the present application provides a rail transit intelligent dispatching method, which is applied to a rail transit intelligent dispatching system, and the method comprises: Obtain the predicted arrival result of the active passenger flow at the target venue; divide the platform space of the rail transit station within the preset distance around the target venue into multiple virtual waiting areas according to the predicted arrival result of the active passenger flow; and configure a departure condition combination library for each virtual waiting area, wherein the departure condition combination library contains multiple sets of preset departure conditions, each set of the preset departure conditions includes a minimum waiting number threshold and a maximum waiting time, and each set of the preset departure conditions corresponds to a scene label, wherein the scene label includes a regular time period and a peak time period; obtain the real-time waiting number and waiting time of each virtual waiting area, The congestion index of the current waiting area is calculated according to the real-time number of waiting passengers and the waiting time; a matching target departure condition is selected from the departure condition combination library according to the congestion index, and a departure instruction is triggered when the number of waiting passengers and the waiting time meet the target departure condition; an operation efficiency score of each virtual waiting area after each departure is calculated; an area with an efficiency score higher than a preset threshold is set as a fast departure area, and an area with an efficiency score lower than the preset threshold is set as a regular departure area, the fast departure area is set with fast departure parameters, and the regular departure area is set with regular departure parameters.

[0007] In the above embodiment, virtual waiting areas are divided according to the results of active passenger flow prediction and a departure condition combination library is configured to avoid excessive gathering of people in local areas of the platform, calculate the congestion index of each waiting area in real time and dynamically trigger the departure instruction, eliminating the problem of long waiting time caused by fixed departure strategy. Based on the operation efficiency score, the fast departure area and the regular departure area are divided, and differentiated departure parameters are used to solve the problem of uneven distribution of transportation resources, making the platform space utilization and passenger flow evacuation more balanced and efficient.

[0008] In combination with some embodiments of the first aspect, in some embodiments, before the step of obtaining the activity passenger flow arrival prediction result of the target venue, dividing the platform space of the rail transit station within a preset distance range around the target venue into multiple virtual waiting areas according to the activity passenger flow arrival prediction result, and configuring a departure condition combination library for each virtual waiting area, the method also includes: establishing an activity feature database and collecting real-time traffic data around the target venue, the activity feature database containing venue capacity, activity type and ticket purchaser residence distribution information, and the real-time traffic data including road congestion status; establishing a venue activity passenger flow prediction model based on the activity feature database and the real-time traffic data; obtaining real-time activity data of the target venue, inputting the real-time activity data into the passenger flow prediction model, and obtaining the activity passenger flow arrival prediction result after the end of the activity, and the real-time activity data including the end time of the activity and on-site number statistics.

[0009] In the above embodiment, an activity feature database containing information on venue capacity, activity type, and the residence distribution of ticket buyers is established, and a venue activity passenger flow prediction model is constructed in combination with traffic data such as real-time road congestion status. The real-time activity data is input into the prediction model to obtain the passenger flow arrival prediction result, so that the system can accurately grasp the passenger flow arrival characteristics in advance, provide accurate data support for the division of the virtual waiting area and the configuration of the departure conditions, and enhance the foresight and accuracy of the scheduling plan.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the step of calculating the operating efficiency score of each virtual waiting area after each departure specifically includes: calculating the average waiting time and the number of evacuees per unit time of each virtual waiting area after each departure, and taking the weighted sum of the average waiting time and the number of evacuees per unit time as the efficiency score of the departure.

[0011] In the above embodiment, the weighted sum of the average waiting time and the number of people evacuated per unit time is used as the efficiency score, and an evaluation mechanism that fully reflects the operating status of the waiting area is established. The efficiency score comprehensively considers the two dimensions of passenger waiting experience and passenger evacuation efficiency, provides a quantitative basis for the division of fast departure areas and regular departure areas, optimizes the platform space resource allocation, and improves the scientific nature of operation scheduling.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after selecting a matching target departure condition from the departure condition combination library according to the congestion index, and triggering the departure instruction when the number of waiting people and the waiting time meet the target departure condition, the method also includes: when the efficiency score is greater than a preset threshold, lowering the minimum waiting number threshold in the corresponding departure condition combination to a preset first number threshold; when the efficiency score is less than the preset threshold, increasing the minimum waiting number threshold in the corresponding departure condition combination to a preset second number threshold.

[0013] In the above embodiment, by adopting the above technical solution, the minimum waiting number threshold in the departure condition combination is dynamically adjusted based on the efficiency score. When the efficiency score is greater than the preset threshold, it is reduced to the preset first number threshold. When the efficiency score is less than the preset threshold, it is increased to the preset second number threshold. An adaptive adjustment mechanism for departure conditions is formed, which enables the departure strategy to always remain in the optimal operating state, thereby improving the stability and reliability of system operation.

[0014] In combination with some embodiments of the first aspect, in some embodiments, after the step of setting the area with an efficiency score higher than a preset threshold as a fast departure area and setting the area with an efficiency score lower than the preset threshold as a regular departure area, the method further includes: dynamically adjusting the partition boundaries of each virtual waiting area based on a change trend of the efficiency score; According to the changes in the boundary of the zone, the distribution density of the waiting crowd is determined, and the operation strategy of the adjacent waiting area is adjusted according to the distribution density of the waiting crowd.

[0015] In the above embodiment, the virtual waiting area zoning boundaries are dynamically adjusted based on the efficiency score change trend, and the adjacent waiting area operation strategy is adjusted according to the waiting crowd distribution density, thus establishing a flexible adjustment mechanism for the waiting area boundaries. The dynamic optimization of the zoning boundaries ensures the load balance of each waiting area, improves the utilization efficiency of platform space resources, and enhances the system's ability to adapt to changes in passenger flow distribution.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of setting the area with an efficiency score higher than a preset threshold as a fast departure area and setting the area with an efficiency score lower than the preset threshold as a regular departure area, the method further includes: adjusting the signal interval and protection partition parameters in the train control equipment in the fast departure area; and establishing a buffer control interval at the junction of the fast departure area and the regular departure area.

[0017] In the above embodiment, the signal interval and protection partition parameters of the train control equipment are adjusted in the fast departure area, and a buffer control section is established at the junction of the fast departure area and the regular departure area, so as to achieve accurate configuration of the train operation control parameters. The setting of the buffer control section ensures a smooth transition between different departure areas and optimizes the safety and continuity of train operation.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of establishing a buffer control interval at the junction of the fast departure area and the regular departure area, the method also includes: setting a parameter adjustment cycle of the train control device according to the efficiency score; within each adjustment cycle, hierarchically configuring the control parameters of the train control device based on the current efficiency score.

[0019] In the above embodiment, the parameter adjustment cycle of the train control equipment is set according to the efficiency score, and the control parameters of the train control equipment are configured in a hierarchical manner based on the current efficiency score in each adjustment cycle, forming a periodic optimization mechanism for the parameters of the train control equipment. The dynamic setting of the parameter adjustment cycle is combined with the hierarchical configuration, so that the train control equipment always works in the optimal control state, realizing the precise and intelligent adjustment of train operation control.

[0020] In a second aspect, an embodiment of the present application provides a rail transit intelligent dispatching system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the rail transit intelligent dispatching system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions. When the above-mentioned computer program product runs on a rail transit intelligent dispatching system, the above-mentioned rail transit intelligent dispatching system executes the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions. When the above instructions are executed on a rail transit intelligent dispatching system, the above rail transit intelligent dispatching system executes the method described in the first aspect and any possible implementation method of the first aspect.

[0023] It is understandable that the rail transit intelligent dispatching system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. This application avoids excessive gathering of people in local areas of the platform by dividing virtual waiting areas according to the results of active passenger flow forecasts and configuring a departure condition combination library. It calculates the congestion index of each waiting area in real time and dynamically triggers departure instructions, eliminating the problem of long waiting time caused by fixed departure strategies. Based on the operation efficiency score, the fast departure area and the regular departure area are divided, and differentiated departure parameters are adopted to solve the problem of uneven distribution of transportation resources, making the platform space utilization and passenger flow evacuation more balanced and efficient.

[0025] 2. This application builds a venue activity passenger flow prediction model by establishing an activity feature database containing venue capacity, activity type, and distribution information of ticket purchasers' residences, combined with real-time road congestion status and other traffic data. Inputting real-time activity data into the prediction model to obtain the passenger flow arrival prediction results enables the system to accurately grasp the passenger flow arrival characteristics in advance, providing accurate data support for the division of virtual waiting areas and the configuration of departure conditions, and enhancing the foresight and accuracy of the scheduling plan.

[0026] 3. This application establishes an evaluation mechanism that comprehensively reflects the operation status of the waiting area by taking the weighted sum of the average waiting time and the number of people evacuated per unit time as the efficiency score. The efficiency score comprehensively considers the two dimensions of passenger waiting experience and passenger evacuation efficiency, provides a quantitative basis for the division of fast departure areas and regular departure areas, optimizes the allocation of platform space resources, and improves the scientific nature of operation scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of the rail transit intelligent dispatching method in the embodiment of the present application; Figure 2 is another flow chart of the rail transit intelligent scheduling method in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a physical device of the rail transit intelligent dispatching system in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] For ease of understanding, the application scenarios of the embodiments of the present application are introduced below.

[0031] When major events are held in large stadiums, spectators often leave in droves after the event. For example, in a stadium that can accommodate 20,000 people, it is estimated that about 12,000 spectators will choose to travel by subway after an important football match. The two nearby subway stations need to evacuate these spectators within 60 minutes. The traditional fixed time interval departure method is difficult to cope with such sudden large passenger flow. Due to the uneven distribution of passenger flow, some platform areas are severely crowded, while other areas are relatively idle, resulting in low efficiency of platform resource utilization. At the same time, the fixed departure interval cannot be dynamically adjusted according to the actual passenger flow changes, resulting in excessively high train load rates during peak hours, long waiting times for passengers, and even safety hazards. Especially under the influence of factors such as weather and emergencies, the fixed operation mode is more difficult to meet the needs of flexible scheduling.

[0032] A subway station uses a passenger flow warning system based on a fixed time window, dividing the entire station into several monitoring areas. The system counts the passenger flow density of each area every 5 minutes, and issues a warning when the density of a certain area exceeds 0.5 people / square meter. After receiving the warning, the station staff will divert the passenger flow through broadcasting and manual guidance. However, this passive response mode has obvious lag, and measures are often taken only after congestion has already formed. At the same time, due to the lack of the ability to predict future passenger flow, it is impossible to adjust the departure plan in advance. For example, when a concert ended, the system began to increase the departure frequency only after detecting that the passenger flow density in area A reached the warning threshold. At this time, the platform was already seriously congested. Although the nearby area B was temporarily idle, it was unable to predict the subsequent passenger flow direction and failed to allocate capacity in time, which ultimately caused the entire evacuation process to last nearly 90 minutes, far exceeding the expected time.

[0033] During a basketball game, the subway station adopting this solution predicts the passenger flow distribution after the game in advance based on the game information, ticket purchase data and real-time traffic conditions. When it is expected that 8,000 spectators will arrive at the subway station within 45 minutes, the system automatically divides the platform into 4 virtual waiting areas and configures differentiated departure conditions for each area. By monitoring the number of people waiting and waiting time in each area in real time, the system dynamically calculates the congestion index. When it is detected that the efficiency score of the north waiting area exceeds 85 points continuously, it is automatically set as a fast departure area, shortening the departure interval to 3 minutes, and adjusting the signal system parameters accordingly. At the same time, a 100-meter buffer zone is set between the fast area and the regular area to achieve a smooth transition of the operation mode. Through this intelligent scheduling strategy, the entire process of the game only takes 40 minutes, and the average waiting time is controlled within 6 minutes, which not only ensures transportation efficiency, but also maintains a good passenger experience. The system also dynamically adjusts the boundaries and operating parameters of each area based on the real-time efficiency score to achieve the optimal configuration of platform resources.

[0034] For ease of understanding, the following describes the process of the method provided by this implementation in combination with the above scenario. Figure 1 , which is a flow chart of the rail transit intelligent scheduling method in an embodiment of the present application.

[0035] S101. Obtain the predicted arrival result of the active passenger flow at the target venue, and divide the platform space of the rail transit station within a preset distance range around the target venue into multiple virtual waiting areas according to the predicted arrival result of the active passenger flow, and configure a departure condition combination library for each virtual waiting area, the departure condition combination library contains multiple groups of preset departure conditions, each group of the preset departure conditions includes a minimum waiting number threshold and a maximum waiting time, and each group of the preset departure conditions corresponds to a scene label, and the scene label includes a regular time period and a peak time period.

[0036] Among them, the event passenger flow arrival prediction result indicates the passenger flow and distribution expected to arrive at the nearby subway station after the target venue event ends; the virtual waiting area refers to several independent waiting areas divided by the platform space according to the passenger flow distribution characteristics; the departure condition combination library is used to store the departure condition parameter sets under different scenarios; the scene label refers to the identification that classifies the departure conditions according to the time period characteristics.

[0037] This step is executed when the system is started and is used to initialize the division of the virtual waiting area and its departure condition configuration. Specifically, the system first obtains the passenger flow forecast data after the target venue event ends, and then divides the platform space of the subway station around the venue into multiple relatively independent waiting areas according to the predicted passenger flow distribution law. Each waiting area is configured with an independent departure condition parameter library. The departure condition parameters include the minimum number of waiting people required and the maximum allowed waiting time. The parameter values ​​are different in different scenarios.

[0038] In some embodiments, the intelligent division of the virtual waiting area can be achieved in the following ways: Optionally, first establish a historical passenger flow distribution heat map, identify high-density gathering areas in the platform space, and use the gathering areas as the core areas of the waiting area; then analyze the interaction between the flow of people in each area and determine the regional boundaries; finally, optimize and adjust the boundaries according to the rationality of the spatial layout. Optionally, use a clustering algorithm to analyze real-time passenger flow data, dynamically identify crowd gathering characteristics, and divide the area in combination with the physical constraints of the platform; determine the number of areas through density clustering; and determine the regional boundaries based on the principle of spatial continuity. It is understandable that other data-driven regional division methods can also be used, which are not limited here.

[0039] S102: Obtain the real-time number of passengers waiting for the bus and the waiting time in each virtual waiting area, and calculate the congestion index of the current waiting area according to the real-time number of passengers waiting for the bus and the waiting time.

[0040] Among them, the real-time number of passengers waiting for the bus refers to the number of passengers currently waiting for the bus in each virtual waiting area; the waiting time refers to the length of time from the passengers entering the waiting area to the current moment; the congestion index is used to characterize the degree of congestion in the waiting area.

[0041] This step is continuously executed during the operation of the system, and is used to monitor and evaluate the operating status of each waiting area in real time. Specifically, the system obtains the number of people waiting in each virtual waiting area in real time through the platform passenger flow detection equipment, and records the time when passengers enter the waiting area, calculates the waiting time, and calculates the congestion index by comprehensively considering the number of people and waiting time.

[0042] In some embodiments, the calculation of the crowding index can be achieved in the following ways: Optionally, first count the density of the number of people waiting in a unit area, and calculate the spatial saturation in combination with the maximum holding density; then use the ratio of the average waiting time to the expected waiting time as the time factor; finally, perform weighted calculation on the spatial saturation and the time factor to obtain the crowding index. Optionally, analyze the standing distance and body posture of the waiting crowd based on image recognition technology to evaluate the comfort level; calculate the fatigue index in combination with the waiting time; and calculate the crowding index by combining the spatial comfort and time fatigue. It is understandable that other calculation methods that objectively reflect the degree of crowding in the waiting area can also be used, which are not limited here.

[0043] S103, selecting a matching target departure condition from the departure condition combination library according to the congestion index, and triggering a departure instruction when the number of waiting persons and the waiting time meet the target departure condition.

[0044] Among them, the congestion index refers to a quantitative indicator that reflects the current congestion status of the waiting area; the target departure condition refers to the optimal departure parameter combination matched according to the current scenario; the departure instruction refers to the departure control signal sent by the system to the train control device. The departure condition matching degree refers to the degree of adaptation between the current congestion index and the departure condition combination. The triggering timing refers to the time point when the system executes the departure instruction when the departure conditions are met.

[0045] This step is executed when the system detects changes in the congestion of the waiting area, and is used to dynamically adjust the departure strategy. Specifically, the system first calculates the matching degree of each condition in the departure condition combination library based on the current congestion index, and selects the condition with the highest matching degree as the target departure condition. Then, the system monitors in real time whether the number of waiting people and the waiting time meet the threshold requirements of the target conditions. When both indicators meet the requirements at the same time, the system immediately sends a departure instruction to the train control device and records the departure time and related operating parameters.

[0046] In some embodiments, intelligent matching and triggering of departure conditions can be achieved in a variety of ways: Optionally, first build a departure condition scoring model, and use factors such as congestion index, waiting time distribution, and crowd density as input variables; then calculate the fitness score of each set of departure conditions, and the condition with the highest score is selected as the target condition; finally, set a dynamic trigger threshold, and trigger departure when the actual indicator exceeds the threshold and continues for a preset period of time. Optionally, establish a departure decision model based on deep reinforcement learning, use the waiting area status as input, and evaluate the benefits of different departure conditions through the value network; use the strategy network to select the optimal departure condition; set a multi-level trigger mechanism, and determine the trigger priority according to the degree of condition satisfaction. It is understandable that other intelligent decision-making methods can also be used to achieve the selection and triggering of departure conditions, which are not limited here.

[0047] S104: Calculate the operation efficiency score of each virtual waiting area after each departure.

[0048] Among them, the operating efficiency score refers to a comprehensive evaluation indicator to measure the departure scheduling effect of the virtual waiting area; the post-departure status refers to the various operating parameters of the waiting area after the train departs; the scoring period refers to the time interval for calculating the efficiency score; the reference benchmark refers to the standard value used for scoring comparison.

[0049] This step is performed after each train departure to evaluate the actual effect of the departure scheduling. Specifically, the system collects various operation data of the waiting area after the train leaves the station, including the change in the number of waiting people, the distribution of waiting time, and the space utilization rate, and calculates the efficiency score of this departure according to the preset scoring model. This score will be used to optimize the departure strategy and regional division in the future.

[0050] In some embodiments, the evaluation and calculation of operating efficiency can be achieved in a variety of ways: optionally, first construct a multi-dimensional evaluation index system, including personnel evacuation rate, average waiting time, spatial balance, etc.; then use the hierarchical analysis method to determine the weight of each indicator; finally, calculate the efficiency score based on the fuzzy comprehensive evaluation method. Optionally, establish a data envelopment analysis model to standardize the input and output indicators of the waiting area operation; calculate the relative efficiency value; and score the efficiency in combination with the historical benchmark value. It is understandable that other scientific efficiency evaluation methods can also be used to quantitatively evaluate the dispatching effect, which is not limited here.

[0051] S105: Set the area with an efficiency score higher than a preset threshold as a fast departure area, and set the area with an efficiency score lower than the preset threshold as a regular departure area. Set fast departure parameters for the fast departure area, and set regular departure parameters for the regular departure area.

[0052] Among them, the efficiency score threshold refers to the critical score value used to distinguish between the fast departure area and the regular departure area; the fast departure area refers to the waiting area with high operating efficiency and high-frequency departure strategy; the regular departure area refers to the waiting area that maintains normal departure frequency; the fast departure parameters are used to define the set of operation control parameters under the high-frequency departure mode, including the shortest departure interval, the minimum number of waiting people, etc.; the regular departure parameters represent the standard operation parameters under the normal departure mode. The regional operation mode refers to the type of waiting area departure scheduling strategy determined by the efficiency score.

[0053] This step is executed after the system completes an evaluation cycle and is used to optimize the regional operation strategy based on the efficiency evaluation results. Specifically, the system first sets the efficiency score classification threshold, and classifies the waiting area above the threshold into the fast departure area, adopts a more aggressive departure strategy, shortens the departure interval, and reduces the minimum waiting number requirement; the waiting area below the threshold is classified into the regular departure area, and the standard departure parameter configuration is maintained. At the same time, the system will establish a parameter adaptive adjustment mechanism to dynamically optimize the departure parameters according to the regional operation status to ensure a balance between operation efficiency and safety.

[0054] In some embodiments, regional classification and parameter configuration can be achieved in a variety of ways: Optionally, first determine the optimal demarcation threshold of the efficiency score based on cluster analysis, and divide the waiting area into two categories: fast and regular; then use pattern recognition methods to analyze the operating characteristics of high-efficiency areas and extract key parameter patterns; finally, build a parameter optimization model, and combine safety constraints to determine the fast departure parameters, including the shortest interval time, the minimum number of departures, etc. Optionally, establish a multi-objective decision model, taking operating efficiency, safety margin, energy consumption and other factors as optimization goals; use genetic algorithms to search for the optimal parameter combination; determine the feasible domain of parameters through simulation verification, and form a parameter configuration plan for each region. It is understandable that other intelligent methods can also be used to achieve classified management and parameter configuration of waiting areas, which are not limited here.

[0055] Before step S101, the method further includes the following steps: An activity feature database is established and real-time traffic data around the target venue is collected. The activity feature database contains information on venue capacity, activity type, and residence distribution of ticket buyers. The real-time traffic data includes road congestion status.

[0056] In this step, the activity feature database refers to a structured data set that stores the attribute information related to the venue activities; the venue capacity indicates the maximum number of people that the venue can accommodate; the activity types include concerts, sports events, exhibitions and other types of activities of different nature; the distribution information of the residence of ticket buyers records the geographical distribution characteristics of the audience; real-time traffic data refers to dynamic information reflecting the operating status of the surrounding road network, and the system establishes a multi-dimensional data collection and storage architecture. At the database level, a relational database is used to store structured information, and a time series database is used to record dynamically changing data. The venue capacity data is obtained by regional statistics: C=∑(Si*Di), where Si is the area of ​​the i-th region and Di is the design density coefficient of the region. The activity type adopts a hierarchical classification system, and the activity attributes are represented by the feature vector [t1, t2,..., tn], including dimensions such as duration, interactivity, and audience age distribution. The distribution of the ticket purchaser's residence is estimated using the kernel density method: f(x, y) = (1 / nh²) ∑ K ((x-xi) / h, (y-yi) / h), where (xi, yi) is the ticket purchaser's location coordinates, K is the kernel function, and h is the bandwidth parameter. Real-time traffic data is collected through intersection monitoring equipment, recording indicators such as vehicle flow, average speed, and queue length, and establishing a minute-level update mechanism.

[0057] Based on the activity feature database and the real-time traffic data, a venue activity passenger flow prediction model is established.

[0058] In this step, the venue activity passenger flow prediction model refers to a mathematical model that predicts the audience's departure behavior based on historical data and real-time information; the prediction results include passenger flow prediction values ​​at different time points and different exits. The system constructs a deep learning prediction model. First, the input data is feature engineered to convert activity features, traffic status, etc. into standardized feature vectors. The LSTM (Long Short-Term Memory Network) structure is used to model the time series features: ht=σ(Wh·[ht-1,xt]+bh), where xt is the input feature at time t and ht is the hidden state. The model also introduces an attention mechanism: α=softmax(W·tanh(V·[h1,...,hn])) to highlight the influence of key time points. Finally, the influence of each feature is integrated through a multi-layer perceptron: y=φ(W2·ReLU(W1·[h,c]+b1)+b2), where h is the LSTM output, c is the context vector, and y is the prediction result. The model training adopts the batch gradient descent method and uses the mean square error as the loss function.

[0059] The real-time activity data of the target venue is obtained, and the real-time activity data is input into the passenger flow prediction model to obtain the passenger flow arrival prediction result after the activity ends. The real-time activity data includes the activity end time and the on-site number of people statistics.

[0060] In this step, real-time activity data refers to dynamic monitoring information during the current activity; the end time of the activity indicates the expected end time of the activity; the on-site headcount refers to the actual number of spectators at the activity site; and the event passenger flow arrival prediction result includes the passenger flow distribution prediction value on the time series. The system executes the real-time prediction process. First, the input data is preprocessed to convert the end time of the activity into a timestamp feature. The number of people on site is obtained through the camera recognition result: N=∑P(Ri), where Ri is the target detection area in the image, and P(Ri) is the number of people in the area. The preprocessed data is input into the prediction model to generate the passenger flow prediction sequence from t+1 to t+k: Q(t+i)=M(X(t), S(t), i), where X(t) is the feature vector at the current moment, S(t) is the model state vector, and i is the prediction step size. The prediction result includes the minute-level passenger flow of each exit, and gives the prediction confidence interval: [Q-1.96σ, Q+1.96σ], where σ is the prediction standard deviation.

[0061] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the rail transit intelligent scheduling method in an embodiment of the present application.

[0062] S201. When the efficiency score is greater than a preset threshold, the minimum waiting number threshold in the corresponding departure condition combination is reduced to a preset first number threshold.

[0063] Among them, the preset threshold refers to the benchmark value for judging the efficiency score, which is usually determined based on historical operation data statistics; the minimum waiting number threshold indicates the minimum number of waiting people required to trigger the departure command; the preset first number threshold indicates the minimum waiting number requirement under high efficiency conditions.

[0064] The system automatically adjusts the departure conditions by monitoring the relationship between the efficiency score and the preset threshold in real time. When the efficiency score exceeds the preset threshold, it means that the current departure strategy is running well, and the system adjusts the minimum waiting number threshold to the preset first number threshold. In the specific adjustment process, first calculate the difference ratio between the efficiency score and the preset threshold, substitute the ratio into the function y=a*exp(-bx)+c (where a, b, c are coefficients obtained by fitting based on historical data) to obtain the threshold adjustment amount. Then determine the new waiting number threshold based on the adjustment amount, and use a progressive adjustment method, with each adjustment not exceeding 15% of the original threshold, to avoid parameter mutations leading to system instability. For example, when the original minimum waiting number threshold is 100 people, if the adjustment amount is calculated to be 25 people, the threshold is gradually adjusted to 80 people in 3 times, with each adjustment interval of no less than 10 minutes.

[0065] S202: When the efficiency score is less than a preset threshold, the minimum waiting number threshold in the corresponding departure condition combination is increased to a preset second number threshold.

[0066] Among them, the preset threshold refers to the benchmark value for determining the efficiency score; the minimum waiting number threshold represents the minimum number of waiting people required for departure; and the preset second number threshold represents the minimum waiting number requirement under low efficiency conditions.

[0067] The system performs a threshold increase operation when it detects that the efficiency score is lower than the preset threshold. First, the threshold adjustment coefficient is calculated based on the reduction in the efficiency score: k=1+α*(TS) / T, where T is the preset threshold, S is the current efficiency score, and α is the adjustment factor. Then multiply the current minimum waiting number threshold by the adjustment coefficient to obtain a new threshold value. If the calculation result exceeds the preset second number threshold, the preset second number threshold is directly used. For example, when the efficiency score drops from 80 points to 60 points and the preset threshold is 70 points, if the current minimum waiting number threshold is 100 people, α is 0.5, then the adjustment coefficient k=1.071, and the new threshold should be adjusted to 107 people. The system checks the new threshold to ensure that it does not exceed the preset second number threshold of 120 people, and finally performs the threshold adjustment operation.

[0068] S203. Dynamically adjust the partition boundaries of each virtual waiting area based on the change trend of the efficiency score.

[0069] Among them, the efficiency score change trend represents the dynamic change law of the efficiency score over a period of time; the partition boundary refers to the dividing line between each virtual waiting area; and dynamic adjustment refers to the process of automatically optimizing the boundary position according to the operating status.

[0070] The system determines the boundary adjustment strategy by analyzing the time series data of efficiency scores. First, the exponential smoothing method is used to predict the short-term efficiency change trend, and the wavelet transform is used to identify the periodic fluctuation characteristics. For adjacent waiting areas, the system calculates the boundary adjustment amount: δ=k*[(E1-E2) / E_avg]*L, where E1 and E2 are the efficiency scores of the adjacent areas, E_avg is the average efficiency score, L is the original boundary length, and k is the adjustment coefficient determined according to the site conditions. When δ is a positive value, the boundary shrinks toward the waiting area with lower efficiency; when δ is a negative value, the boundary expands toward the waiting area with higher efficiency. When performing boundary adjustment, the system calculates the area ratio and shape parameters of each area in real time, and ensures that the adjusted zoning layout meets the actual operation needs through constraints. For example, the area of ​​any waiting area shall not be less than the minimum area requirement, the shape of the area shall not be seriously distorted, and sufficient buffer width must be reserved between adjacent areas.

[0071] S204: Determine the distribution density of the waiting crowd according to the change of the partition boundary, and adjust the operation strategy of the adjacent waiting area according to the distribution density of the waiting crowd.

[0072] Among them, the change of partition boundary represents the dynamic adjustment result of the boundary position of the virtual waiting area; the distribution density of the waiting crowd refers to the number of people waiting per unit area; the operation strategy includes a combination of operating parameters such as departure frequency, station entry guidance, and platform service.

[0073] The system calculates the crowd distribution characteristics of each area in real time according to the adjustment of the partition boundary. First, the density function model is used to analyze the crowd distribution: D(x, y) = ∑ [Pi * exp (- ((x-xi) ² + (y-yi) ²) / σ ²)], where (x, y) is the platform coordinate point, (xi, yi) is the i-th crowd gathering center, Pi is the weight of the number of people in the center, and σ is the distribution range parameter. Based on the density distribution results, the system calculates the congestion index of each area: C = ∫∫D (x, y) dxdy / A, where A is the area of ​​the area. When the difference in congestion between adjacent waiting areas exceeds the preset threshold, the system automatically adjusts the operation strategy. For areas with high congestion, passenger flow is dispersed by increasing the departure frequency, optimizing the allocation of entry channels, and increasing the number of platform guides. At the same time, adjacent low-congestion areas adopt passenger flow guidance strategies to reasonably guide some passengers to the surrounding waiting areas to achieve balanced utilization of platform resources.

[0074] S205. Adjust the signal interval and protection zone parameters in the train control equipment in the fast departure area.

[0075] Among them, train control equipment refers to the signal system device that controls the operation of trains; the signal interval indicates the minimum safe distance between adjacent trains; the protection zone parameters include zone length, protection level, reaction time and other safety control parameters.

[0076] The system optimizes the parameters of train control equipment according to the operating characteristics of the fast departure area. The signal interval adjustment adopts a dynamic interval calculation model: I=√(2aL / b+v²)+S, where L is the train length, v is the running speed, a is the acceleration, b is the braking deceleration, and S is the safety margin. The protection partition parameter adjustment is based on the risk assessment results, and the Markov chain model is used to calculate the state transition probability of each partition to determine the key protection parameters. During the specific adjustment process, the system first reduces the signal interval time, shortening the standard 5-minute interval to 3 minutes; secondly, it compresses the length of the protection partition, reducing the standard partition length of 600 meters to 450 meters while ensuring safety; finally, it optimizes the signal processing logic, reducing the conventional 2-second processing delay to 1 second, and improving the system response speed.

[0077] S206: Establish a buffer control area at the junction of the fast departure area and the regular departure area.

[0078] Among them, the buffer control interval refers to the transition area set up between the fast departure area and the regular departure area; the junction indicates the dividing line between two different operating modes; the control parameters include technical indicators such as interval length, operating speed, and signal configuration.

[0079] The system establishes a buffer control mechanism at the junction of the two types of areas. The length between the buffers is calculated by a dynamic programming algorithm: L=max{VsTs, VfTf}, where Vs and Vf are standard speed and fast speed respectively, and Ts and Tf are the corresponding coordination times. The system sets a speed gradient zone between the buffers, and the train speed gradually transitions according to V(x)=V0+(V1-V0)*[1-exp(-kx / L)], where V0 and V1 are the speed values ​​at both ends, x is the position coordinate in the buffer, and k is the smoothing coefficient. At the same time, a hierarchical control strategy is implemented, and a three-level protection mechanism is set in the buffer: early warning protection, emergency protection and safety protection to ensure that the train completes the operation mode switch safely and smoothly. For example, when a train in the fast departure area enters the buffer zone, the system automatically reduces its operating speed, gradually reducing the speed from 90km / h to 60km / h, and smoothly transitioning to the normal operation state.

[0080] S207. Setting a parameter adjustment period of the train control device according to the efficiency score.

[0081] Among them, the parameter adjustment cycle refers to the update time interval of the control parameters of the train control equipment; the efficiency score refers to the comprehensive evaluation index reflecting the system operation effect; the train control equipment parameters include technical parameters such as signal control time, protection distance, and response delay.

[0082] The system determines the parameter adjustment cycle based on the fluctuation characteristics of the efficiency score. First, the autocorrelation analysis method is used to calculate the periodic characteristics of the efficiency score: R(τ)=E[(St-μ)(St+τ-μ)] / σ², where St is the efficiency score at time t, τ is the time delay, and μ and σ² are the mean and variance, respectively. Then the benchmark cycle T0 is determined based on the maximum autocorrelation coefficient. The actual adjustment cycle is calculated by the function T=T0*[1+β*(Sc-S0) / S0], where Sc is the current efficiency score, S0 is the benchmark efficiency score, and β is the adjustment coefficient. When the efficiency score is high, the system shortens the adjustment cycle to maintain a good operating state; when the efficiency score is low, the adjustment cycle is extended to maintain system stability. For example, when the benchmark cycle is 30 minutes and the current efficiency score is 20% higher than the benchmark, the actual adjustment cycle will be shortened to 25 minutes, increasing the frequency of parameter optimization.

[0083] S208. In each adjustment cycle, control parameters of the train control device are configured in a hierarchical manner based on the current efficiency score.

[0084] Among them, hierarchical configuration refers to setting different levels of control parameters according to efficiency scores; control parameters include operation control variables such as signal interval time, protection zone length, speed limit, etc.; the current efficiency score represents the operation evaluation result within this cycle.

[0085] The system performs hierarchical parameter configuration at the beginning of each adjustment cycle. First, the mapping relationship between efficiency score and parameter is constructed: P=P0*[1+α*f(S)], where P is the actual parameter value, P0 is the standard parameter value, S is the efficiency score, and f(S) is a piecewise function: when S≥90, f(S)=-0.2; when 70≤S<90, f(S)=-0.1; when 50≤S<70, f(S)=0; when S<50, f(S)=0.1. α is the adjustment coefficient related to the parameter type. The system calculates the configuration values ​​for different types of control parameters, such as signal interval time, protection partition length, etc. For example, when the efficiency score is 95 points, the system shortens the standard 5-minute signal interval to 4 minutes, and reduces the 600-meter protection partition length to 480 meters. Through this hierarchical configuration mechanism, the system realizes the refined management of control parameters to ensure the balance between equipment operation efficiency and safety.

[0086] The rail transit intelligent dispatching system in the embodiment of the present invention is described below from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the structure of a physical device of the rail transit intelligent dispatching system in an embodiment of the present application.

[0087] It should be noted that Figure 3 The structure of the rail transit intelligent dispatching system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0088] like Figure 3 As shown, the rail transit intelligent dispatching system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0089] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0090] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0091] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0092] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0093] Specifically, the rail transit intelligent dispatching system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the rail transit intelligent dispatching method provided in the above embodiment is implemented.

[0094] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the rail transit intelligent dispatching system described in the above embodiment; or may exist independently without being assembled into the rail transit intelligent dispatching system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the rail transit intelligent dispatching system, the rail transit intelligent dispatching system implements the rail transit intelligent dispatching method provided in the above embodiment.

[0095] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0096] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0097] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A rail transit intelligent scheduling method, characterized in that: Applied to the rail transit intelligent dispatching system, the method comprises: Obtain the predicted arrival result of the active passenger flow at the target venue, and divide the platform space of the rail transit station within a preset distance around the target venue into multiple virtual waiting areas according to the predicted arrival result of the active passenger flow, and configure a departure condition combination library for each virtual waiting area, wherein the departure condition combination library includes multiple groups of preset departure conditions, each group of the preset departure conditions includes a minimum waiting number threshold and a maximum waiting time, and each group of the preset departure conditions corresponds to a scene label, and the scene label includes a regular time period and a peak time period; Obtaining the real-time number of people waiting for buses and the waiting time in each virtual waiting area, and calculating the congestion index of the current waiting area according to the real-time number of people waiting for buses and the waiting time; Selecting a matching target departure condition from the departure condition combination library according to the congestion index, and triggering a departure instruction when the number of waiting persons and the waiting time meet the target departure condition; Calculating the operation efficiency score of each virtual waiting area after each departure; The area with an efficiency score higher than a preset threshold is set as a fast departure area, and the area with an efficiency score lower than the preset threshold is set as a regular departure area. The fast departure area is set with fast departure parameters, and the regular departure area is set with regular departure parameters.

2. The method according to claim 1, characterized in that Before the step of obtaining the predicted arrival result of the active passenger flow at the target venue, dividing the platform space of the rail transit station within a preset distance range around the target venue into a plurality of virtual waiting areas according to the predicted arrival result of the active passenger flow, and configuring a departure condition combination library for each virtual waiting area, the method further includes: Establishing an activity feature database and collecting real-time traffic data around the target venue, wherein the activity feature database includes information on venue capacity, activity type, and residence distribution of ticket buyers, and the real-time traffic data includes road congestion status; Establishing a venue activity passenger flow prediction model based on the activity feature database and the real-time traffic data; The real-time activity data of the target venue is obtained, and the real-time activity data is input into the passenger flow prediction model to obtain the passenger flow arrival prediction result of the activity after the activity ends. The real-time activity data includes the activity end time and the on-site number of people statistics.

3. The method according to claim 1, characterized in that The step of calculating the operation efficiency score of each virtual waiting area after each departure specifically includes: The average waiting time and the number of people evacuated per unit time of each virtual waiting area after each departure are calculated, and the weighted sum of the average waiting time and the number of people evacuated per unit time is used as the efficiency score of the departure.

4. The method according to claim 1, characterized in that After selecting a matching target departure condition from the departure condition combination library according to the congestion index and triggering a departure instruction when the number of waiting persons and the waiting time meet the target departure condition, the method further includes: When the efficiency score is greater than a preset threshold, the minimum waiting number threshold in the corresponding departure condition combination is reduced to a preset first number threshold; When the efficiency score is less than a preset threshold, the minimum waiting number threshold in the corresponding departure condition combination is increased to a preset second number threshold.

5. The method according to claim 1, characterized in that After the step of setting the area with an efficiency score higher than the preset threshold as a fast departure area and setting the area with an efficiency score lower than the preset threshold as a regular departure area, the method further includes: Based on the changing trend of the efficiency score, dynamically adjust the partition boundaries of each virtual waiting area; According to the change of the partition boundary, the distribution density of the waiting crowd is determined, and the operation strategy of the adjacent waiting area is adjusted according to the distribution density of the waiting crowd.

6. The method according to claim 1, characterized in that After the step of setting the area with an efficiency score higher than the preset threshold as a fast departure area and setting the area with an efficiency score lower than the preset threshold as a regular departure area, the method further includes: adjusting the signal interval and protection partition parameters in the train control equipment in the fast departure area; A buffer control section is established at the junction of the fast departure area and the regular departure area.

7. The method according to claim 6, characterized in that After the step of establishing a buffer control zone at the junction of the fast departure area and the regular departure area, the method further includes: Setting a parameter adjustment period of the train control device according to the efficiency score; In each adjustment cycle, the control parameters of the train control device are configured in a hierarchical manner based on the current efficiency score.

8. A rail transit intelligent dispatching system, characterized in that: The rail transit intelligent scheduling system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the rail transit intelligent scheduling system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the rail transit intelligent dispatching system, the rail transit intelligent dispatching system executes the method as described in any one of claims 1-7.

10. A computer program product, characterized in that When the computer program product runs on a rail transit intelligent dispatching system, the rail transit intelligent dispatching system executes the method as described in any one of claims 1 to 7.

Images