A hierarchical collaborative control system and method for virtual marshaling trains

Abstract

The present invention discloses a hierarchical collaborative control system and method for virtual marshaling trains, which belongs to the field of rail transit signal control technology. The electronic map processing module queries the track line information of the entire train operation according to the train operation plan information to form the track line operation data of the virtual marshaling train. The recommended driving curve planning module regards the virtual marshaling train as a whole according to the track line operation data of the virtual marshaling train, and collaboratively plans the recommended driving curves of all train units in the virtual marshaling train. The real-time tracking control module controls the train units to operate according to their respective recommended driving curves based on the track line operation data of the virtual marshaling train, thereby achieving the expected goal of synchronous operation of the virtual marshaling train. The present invention takes into account the goal of overall synchronous operation of the virtual marshaling train, uniformly plans and manages the control behaviors of all train units, and greatly improves the tracking efficiency of the virtual marshaling train.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit signal control, and in particular to a layered collaborative control system and method for virtual marshaled trains. Background Art

[0002] With the continuous acceleration of economic development and urbanization, urban traffic congestion has become increasingly serious. At the same time, due to the expansion of urban areas, the problem of traffic connectivity between central urban areas and suburban areas has become increasingly serious. As an important means to alleviate traffic pressure and meet the travel needs of urban residents, the construction of urban rail transit (hereinafter referred to as "urban rail") has achieved remarkable results in recent years. The rapid development of the scale of urban rail transit networks, the uneven temporal and spatial distribution of passenger flow, and the increasingly prominent irregular dynamic changes have also put forward higher requirements for the further optimization of vehicle and line resources and the matching degree of capacity and volume.

[0003] In response to the above needs, virtual coupling (VC) technology is a widely recognized solution. Virtual coupling technology can greatly shorten the running distance of train units that are not physically coupled, allowing them to provide transportation operation services like physically coupled trains. Virtual coupling technology can realize online, dynamic and flexible adjustment of vehicle configuration and marshaling methods, thereby improving the effective utilization rate of vehicle and line resources, which can not only meet the large capacity demand during peak passenger flow, but also reduce the vehicle empty rate during off-peak and off-peak periods. Therefore, by developing virtual marshaling technology, it is possible to safely and efficiently control train units to execute online dynamic marshaling and disassembly as planned, and maintain stable operation and synchronous control at small intervals in a virtual marshaling manner. This can reduce train operation energy consumption and save transportation costs without reducing service quality, which is of great significance to the green and sustainable development of urban rail transit.

[0004] At present, the operation control system and method for virtual marshaling trains do not actually consider the integrity and synchronization of the operation of virtual marshaling trains. They are only based on the idea of ​​multi-car tracking operation at a smaller safety protection distance. The pilot train runs independently according to its own plan, and the follower train tracks the pilot train according to the smaller safety protection distance. In addition, the operation control system architecture does not focus on the synchronous operation of virtual marshaling trains. The train unit still uses the original system architecture of single-car planning and control. Therefore, it is inevitable that the virtual marshaling train will run asynchronously from the system level. In short, the existing operation control system and method do not reflect the idea that "the virtual marshaling train is a whole, and all train units need to operate in a coordinated and consistent manner to jointly achieve the operation goals", and this operation control system and method have the following problems:

[0005] 1. A following train that simply tracks the lead train results in low operational efficiency and difficulty achieving operational targets. Existing following trains mostly use simplified tracking spacing (e.g., fixed spacing, fixed time intervals, etc.) to track the lead train. This makes it impossible to track the lead train at the minimum spacing permitted by safety conditions, resulting in reduced tracking efficiency for virtual train formations.

[0006] 2. The following train tracks the real-time status of the lead train, which places extremely high demands on communication transmission between the two trains. At the same time, because virtual marshaling trains use more sophisticated and complex safety distance calculation methods, the performance requirements for real-time calculation of train units are also high. Various random environmental interferences can easily lead to deviations in tracking operations, or even dangers.

[0007] 3. The platform stop times between virtual train units vary significantly. The lead train operates independently according to its own plan, without considering the integrity and synchronization of the virtual train operation. (For example, when the track speed limit changes from low to high, the lead train will accelerate directly after clearing the low-speed limit section, while the following train is still in the low-speed section, resulting in an increase in the distance between the two trains. The uncoordinated operation of the two trains also leads to a large time difference when stopping at the platform.)

[0008] Virtual marshaling technology does not simply break through the absolute braking distance between trains to allow trains to track closer, but rather enables multiple train units to form a unified virtual train based on a closer tracking distance. This new type of virtual marshaling train is essentially regarded as a single vehicle during operation, and the units of the virtual marshaling train cooperate with each other, rather than a simple relationship of a follower car chasing a lead car. Summary of the Invention

[0009] The purpose of the present invention is to provide a hierarchical collaborative control system and method for virtual marshaling trains, which takes into account the goal of overall synchronous operation of virtual marshaling trains, uniformly plans and manages the control behaviors of all train units, and greatly improves the tracking efficiency of virtual marshaling trains.

[0010] To achieve the above object, the present invention provides the following solutions:

[0011] A virtual train hierarchical collaborative control system, characterized by comprising: an information interaction module, an electronic map processing module, a recommended driving curve planning module and multiple real-time tracking control modules;

[0012] One virtual marshaling train unit corresponds to one real-time tracking control module;

[0013] The information interaction module is connected to the electronic map processing module; the information interaction module is used to obtain train operation plan information from the train automatic operation monitoring system and output the train operation plan information to the electronic map processing module;

[0014] The electronic map processing module is connected to the recommended driving curve planning module and the multiple real-time tracking control modules respectively; the electronic map processing module is used to store track line information, and query the track line information of the entire train operation according to the train operation plan information, to form the track line operation data of the virtual marshaling train, and output it to the recommended driving curve planning module and the multiple real-time tracking control modules at the same time;

[0015] The recommended driving curve planning module is connected to multiple real-time tracking control modules; the recommended driving curve planning module is used to regard the virtual marshaling train as a whole based on the track line operation data of the virtual marshaling train, coordinately plan the recommended driving curves of all train units in the virtual marshaling train, and send the recommended driving curves to the real-time tracking control module of each train unit;

[0016] The real-time tracking control module is used to control the train units to operate according to their respective recommended driving curves based on the track line operation data of the virtual marshaling train, so as to achieve the expected goal of synchronous operation of the virtual marshaling train.

[0017] Optionally, the train operation plan information includes: train operation destination, expected operation time and temporary speed limit in section;

[0018] The track line information includes: track line logical section number, stop point, switch, transponder, slope and track static speed limit.

[0019] A layered collaborative control method for a virtual marshaled train, comprising:

[0020] Establish a mathematical model for optimizing the coordinated operation of virtual trains between stations;

[0021] Determine the track line operation data between virtual marshaling train stations based on the train station operation plan information;

[0022] Solving the inter-station collaborative operation optimization mathematical model of the virtual marshaling train based on the track line operation data between the virtual marshaling train stations to obtain the recommended inter-station driving curve for each train unit in the virtual marshaling train;

[0023] According to the real-time status of each train unit and the real-time status of the train units before and after each train unit, based on the track line operation data between the virtual marshaling train stations, the recommended driving curves between the stations of each train unit are tracked in real time, and ultimately the expected goal of synchronous operation of the virtual marshaling train is achieved.

[0024] Optionally, the mathematical model for optimizing the coordinated operation of virtual train stations is:

[0025]

[0026] stv i (k)≤v EBI (s i (k))

[0027] u min ≤u i (k)≤u max

[0028] j min ≤u i (k+1)-u i (k)≤j max

[0029] d(k)≤s1(k)-L-s2(k)

[0030] v i (k)≤V

[0031] i=1,2

[0032] k=1,2,…,n-1,n

[0033] Among them, Q1, Q2, Q3, and Q4 are the optimization weight coefficients of precise parking, punctual operation, synchronous departure, and synchronous parking indicators respectively. and To plan and simulate parking locations for the pilot train and the following train, and The actual target stopping points for the lead train and the following train. and To simulate the stopping time of the pilot train and the following train, and is the initial departure time of the lead train and the following train, T g Establishing operating times between stations; represents the interval running time of the planning simulation; i = 1, 2 are the pilot train and the following train respectively; n represents the total running time of the virtual marshaling train between stations; u i (k),u i (k+1) represents the acceleration controlled by the train unit at time k and time k+1; U i represents the set of controlled accelerations of the train unit at each moment when it is running between stations; s1(k) and s2(k) represent the head positions of the lead train and the following train at time k, respectively; v i (k) represents the speed of the train unit at time k; u min and umax Indicates the train unit control acceleration limit value; j min and j max represents the limit value of the train unit impact rate; d(k) represents the minimum tracking distance between trains; L is the length of the train unit; V represents the speed limit of the track line, v EBI (s i (k)) represents the emergency braking intervention speed at time k.

[0034] Optionally, based on the train station operation plan information, determining the track line operation data between the virtual marshaling train stations specifically includes:

[0035] Obtaining electronic map information of the track line; the electronic map information includes a plurality of logical segments obtained by dividing the track line, and track line data for each logical segment; the track line data includes a speed limit start point, a speed limit end point, a slope start point, a slope end point, a curve radius start point, and a curve radius end point;

[0036] All logical sections included in the operation plan stations and the track line data of each logical section are searched from the electronic map information as the track line operation data between the virtual marshaling train stations.

[0037] Optionally, based on the track line operation data between virtual marshaling train stations, solving the inter-station collaborative operation optimization mathematical model of the virtual marshaling train to obtain the recommended driving curve between stations for each train unit in the virtual marshaling train specifically includes:

[0038] Based on the track line operation data between virtual marshaling train stations, a sequential quadratic programming method, an active set method, a heuristic algorithm or a reinforcement learning algorithm is used to solve the mathematical model for the coordinated operation optimization between the virtual marshaling train stations, and a control acceleration sequence for the entire virtual marshaling train operation plan between stations is obtained;

[0039] According to the control acceleration sequence, the train dynamics equation is used to obtain the position, velocity, and acceleration of the entire inter-station operation plan of the train unit, thereby determining the recommended driving curve between stations for each train unit in the virtual marshaling train; the train dynamics equation is:

[0040] a i (k+1)=u i (k)+g i (s i (k),v i (k));

[0041] v i (k+1)=v i (k)+τa i (k);

[0042]

[0043] Among them, a i (k), a i (k+1) represents the total acceleration at time k and time k+1, v i (k), v i (k+1) represents the speed at time k and time k+1 respectively, s i (k), s i (k+1) represents the position at time k and time k+1 respectively, τ represents the time calculation step, g i (s i (k),v i (k)) represents the function of running resistance and the current position and speed of the train unit.

[0044] Optionally, according to the real-time status of each train unit and the real-time status of the train units preceding and following each train unit, based on the track line operation data between the virtual marshaling train stations, the recommended driving curve between the stations of each train unit is tracked in real time, specifically including:

[0045] Based on the track line operation data between virtual marshaling train stations, a full-time and space safety protection method is used to determine the emergency braking intervention speed of the train unit at any moment;

[0046] According to the real-time status of each train unit and the real-time status of the train units before and after each train unit, the feedforward PID control algorithm is used to track the recommended driving curve between stations of each train unit in real time, and the real-time tracking speed of the train unit is controlled to be less than or equal to the emergency braking intervention speed.

[0047] Optionally, the calculation formula of the feedforward PID control algorithm is:

[0048]

[0049] Where, To actually run k n The control acceleration command at the moment, K P , K I , K D are the parameters of proportional, integral and differential links respectively, e(k n ) is the actual train speed v(k n ) and recommended driving curve speed v r (k n ), e(k n )=v(k n )-v r (k n );∫e(k n ) is the actual train position s(k n) and recommended driving curve position s r (k n ), ∫e(k n )=s(k n )-s r (k n ); is the actual train acceleration a(k n ) and recommended driving curve acceleration a r (k n ), u adj Adjust the control instruction value for the spacing adjustment part.

[0050] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0051] The present invention discloses a hierarchical collaborative control system and method for virtual marshaling trains. An electronic map processing module queries track information for the entire train operation according to train operation plan information to form track operation data for the virtual marshaling train. A recommended driving curve planning module, based on the track operation data of the virtual marshaling train, treats the virtual marshaling train as a whole and collaboratively plans recommended driving curves for all train units in the virtual marshaling train. A real-time tracking control module, based on the track operation data of the virtual marshaling train, controls the train units to operate according to their respective recommended driving curves, thereby achieving the expected goal of synchronized operation of the virtual marshaling train. The present invention takes into account the goal of synchronized operation of the virtual marshaling train as a whole, uniformly plans and manages the control behaviors of all train units, and greatly improves the tracking efficiency of the virtual marshaling train. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 A schematic diagram of the structure of a hierarchical collaborative control system for a virtual marshaled train provided by an embodiment of the present invention;

[0054] Figure 2 A schematic diagram of a layered collaborative control method for a virtual marshaled train provided in an embodiment of the present invention;

[0055] Figure 3 A flowchart of a layered collaborative control method for a virtual marshaled train provided by an embodiment of the present invention;

[0056] Figure 4A structural block diagram of a feedforward PID controller provided in an embodiment of the present invention;

[0057] Figure 5 A schematic diagram of a speed-time curve provided by an embodiment of the present invention;

[0058] Figure 6 A schematic diagram of a speed-position curve provided by an embodiment of the present invention;

[0059] Figure 7 A schematic diagram of a spacing-time curve provided in an embodiment of the present invention;

[0060] Figure 8 A schematic diagram of an acceleration-time curve provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] Example 1

[0064] The innovation of the present invention lies in proposing a hierarchical control system architecture for the synchronous operation of virtual marshaling trains, including a planning layer and a control layer. This hierarchical control system takes into account the goal of overall synchronous operation of virtual marshaling trains from the architectural design level, and uniformly plans and manages the control behaviors of all train units, thus solving the problem of asynchronous operation of virtual marshaling trains from a system level.

[0065] like Figure 1 As shown, an embodiment of the present invention provides a virtual train hierarchical collaborative control system, comprising: an information interaction module, an electronic map processing module, a recommended driving curve planning module, and multiple real-time tracking control modules. One virtual train unit corresponds to one real-time tracking control module.

[0066] The information interaction module is connected to the electronic map processing module; the information interaction module is used to obtain train operation plan information from the train automatic operation monitoring system and output the train operation plan information to the electronic map processing module.

[0067] The electronic map processing module is connected to the recommended driving curve planning module and multiple real-time tracking control modules respectively; the electronic map processing module is used to store track line information, and query the track line information of the entire train operation according to the train operation plan information, form the track line operation data of the virtual marshaling train, and output it to the recommended driving curve planning module and multiple real-time tracking control modules at the same time.

[0068] The recommended driving curve planning module is connected to multiple real-time tracking control modules; the recommended driving curve planning module is used to regard the virtual marshaling train as a whole based on the track line operation data of the virtual marshaling train, coordinately plan the recommended driving curves of all train units in the virtual marshaling train, and send the recommended driving curves to the real-time tracking control module of each train unit.

[0069] The real-time tracking control module is used to control the train units to operate according to their respective recommended driving curves based on the track line operation data of the virtual marshaling train, so as to achieve the expected goal of synchronous operation of the virtual marshaling train.

[0070] Among them, the driving curve planning module is recommended as the planning layer, and the real-time tracking control module is recommended as the control layer.

[0071] Specifically, the system consists of the following modules:

[0072] 1) Information Interaction Processing Module: This module handles information interaction with the Automatic Train Supervision (ATS) system or the trackside resource control system. This module receives information such as the train's destination, expected travel time, and temporary speed limits. It then outputs this information to the electronic map storage and processing module, the recommended driving curve planning module, and the real-time tracking control module.

[0073] 2) Electronic Map Storage and Processing Module: This module stores route information, such as track logical link numbers, stops, switches, transponders, slopes, and static speed limits. Based on the train operation plan information received from the Information Interaction Processing Module, this information is converted into mathematical variables for use by other modules and output to the Recommended Driving Curve Planning Module and the Real-Time Tracking Control Module.

[0074] 3) Recommended Driving Curve Planning Module: This module plans recommended driving curves for all train units within a virtual train. Based on the train's destination, expected travel time, and other received operating conditions, this module considers relevant electronic map information and collaboratively plans the virtual train as a whole, requiring all train units to work together to achieve the same transportation goal (e.g., ensuring the virtual train reaches its destination in the shortest possible time). This module sends the resulting recommended driving curves to the real-time tracking control module of each train unit.

[0075] 4) Real-time tracking control module: This module is used to control the virtual train unit to operate in the desired manner. The specific control logic will be described in the subsequent control method section.

[0076] The system centrally plans and manages the operational control of all virtual train units. Based on the operational plan and line conditions, the planning layer generates recommended driving curves for coordinated operation of all train units, eliminating the need for chase trains. The control layer tracks the recommended driving curves for actual train control, offloading the burden of real-time tracking calculations from the following trains. This layered collaborative control ensures that virtual trains meet the goal of synchronized operation within the section.

[0077] Example 2

[0078] The embodiment of the present invention provides a layered cooperative control method for a virtual train formation, which can be applied to the layered cooperative control system of the virtual train formation in embodiment 1. Figure 2 and Figure 3 , the hierarchical collaborative control method includes:

[0079] Step 1: Establish a mathematical model for optimizing the coordinated operation of virtual trains between stations.

[0080] The planning layer receives the converted electronic map information. The optimal control problem is established. The inter-station optimization problem is constructed as follows:

[0081] ① Optimization goals include precise parking, punctual operation, synchronized departure and synchronized parking.

[0082] ② Constraints include comfort constraints, vehicle traction / braking characteristics constraints, safe tracking distance constraints, and track line speed limit constraints.

[0083] ③The decision variable is set as the control acceleration of each train unit.

[0084] The optimization problem of coordinated operation between virtual marshaling train stations can be expressed in the following mathematical form:

[0085]

[0086] stv i (k)≤v EBI (s i (k))

[0087] u min ≤u i (k)≤u max

[0088] j min ≤u i (k+1)-ui (k)≤j max

[0089] d(k)≤s1(k)-L-s2(k)

[0090] v i (k)≤V

[0091] i=1,2

[0092] k=1,2,…,n-1,n

[0093] Among them, Q1, Q2, Q3, and Q4 are the optimization weight coefficients of precise parking, punctual operation, synchronous departure, and synchronous parking indicators respectively. and To plan and simulate parking locations for the pilot train and the following train, and The actual target stopping points for the lead train and the following train. and To simulate the stopping time of the pilot train and the following train, and is the initial departure time of the lead train and the following train, T g Establishing operating times between stations; represents the interval running time of the planning simulation; i = 1, 2 are the pilot train and the following train respectively; n represents the total running time of the virtual marshaling train between stations; u i (k),u i (k+1) represents the acceleration controlled by the train unit at time k and time k+1; U i represents the set of controlled accelerations of the train unit at each moment when it is running between stations; s1(k) and s2(k) represent the head positions of the lead train and the following train at time k, respectively; v i (k) represents the speed of the train unit at time k; u min and u max Indicates the train unit control acceleration limit value; j min and j max represents the limit value of the train unit impact rate; d(k) represents the minimum tracking distance between trains; L is the length of the train unit; V represents the speed limit of the track line, v EBI (s i (k)) represents the emergency braking intervention speed at time k.

[0094] Step 2: According to the train station operation plan information, determine the track line operation data between the virtual marshaling train stations.

[0095] The information interaction processing module obtains the interval operation plan and index requirements of the virtual marshaling train from the train automatic operation monitoring system and passes the information to the electronic map storage and processing module. The module then extracts and processes the track line data for the virtual marshaling train's planned operation. The processing process is as follows:

[0096] Track lines are divided into logical segments (links). The electronic map storage and processing module contains the initial location and length of each logical segment. Track speed limits, slopes, curve radii, and transponder placement information are given as logical segment numbers plus offsets. By querying the initial location of a corresponding logical segment number and adding the corresponding offset, the corresponding track line information can be determined. Track line data for the entire route is queried based on the train operation plan and used by the recommended driving curve planning module and the real-time tracking control module.

[0097] Step 3: Based on the track line operation data between the virtual marshaling train stations, solve the mathematical model for the coordinated operation optimization between the virtual marshaling train stations to obtain the recommended driving curve between the stations of each train unit in the virtual marshaling train.

[0098] For the optimization problem in step 1, a general solution algorithm, such as sequential quadratic programming, active set method, heuristic algorithm, or reinforcement learning algorithm, is used to obtain the control acceleration sequence for the entire inter-station operation of the virtual marshaling train. By applying the control acceleration sequence to the train dynamics equation, the position, speed, and acceleration data of the train unit throughout the operation can be obtained, i.e., the recommended driving curve for the virtual marshaling train. The train dynamics equation has the following form:

[0099] a i (k+1)=u i (k)+g i (s i (k),v i (k));

[0100] v i (k+1)=v i (k)+τa i (k);

[0101]

[0102] Among them, a i (k), a i (k+1) represents the total acceleration at time k and time k+1, v i (k), v i (k+1) represents the speed at time k and time k+1 respectively, s i (k), s i(k+1) represents the position at time k and time k+1 respectively, τ represents the time calculation step, g i (si(k),v i (k)) represents the function of running resistance and the current position and speed of the train unit.

[0103] At this point, the inter-station operation strategy of the virtual marshaling train has been obtained. The next step is to ensure that the train units can run according to the recommended driving curve to achieve the inter-station operation index requirements.

[0104] Step 4: Based on the real-time status of each train unit and the real-time status of the train units before and after each train unit, and based on the track line operation data between the virtual marshaling train stations, the recommended driving curves between the stations of each train unit are tracked in real time, and the expected goal of synchronous operation of the virtual marshaling train is finally achieved.

[0105] The detailed process is as follows: based on the track line operation data between virtual marshalling train stations, a full-time and space safety protection method is used to determine the emergency braking intervention speed of the train unit at any moment; according to the real-time status of each train unit and the real-time status of the train units before and after each train unit, a feedforward PID control algorithm is used to track the recommended driving curve between stations of each train unit in real time, and the real-time tracking speed of the train unit is controlled to be less than or equal to the emergency braking intervention speed.

[0106] The control layer calculates the control action of the train unit according to the specific control objectives and the corresponding control logic and applies it to the train. Under the hierarchical collaborative control system of the virtual marshaling train, the control goal of each train unit is to track the recommended driving curve. The planning layer sends the available recommended driving curve to the control layer of each train unit in the virtual marshaling train. The control layer of each train unit calculates the control action based on its own recommended driving curve and the real-time status of the vehicle. At the same time, each train unit will also receive the real-time status information of the front and rear trains and make adjustments based on the relative operating relationship (for example, if the actual distance between the two vehicles is closer than the planned distance, the distance between the two vehicles will be increased by adjusting their respective controls). General control algorithms such as PID control and model predictive control can achieve real-time tracking of the recommended driving curve. Here we design a feedforward PID control algorithm for curve tracking. The controller structure block diagram is shown below. Figure 4 shown.

[0107] The control acceleration output by the controller consists of three parts: a feedforward part that controls the acceleration based on the recommended driving curve, a conventional PID controller part, and an adjustment part based on the distance between the two train units. The feedforward part is the average value of the control acceleration of the 10 recommended driving curves from the current moment onward; the PID controller part adjusts the proportional, integral, and differential parameters K according to the error between the actual train state and the recommended driving curve. P ,KI ,K D To output control instructions; the distance adjustment part will adjust the control instructions appropriately when it detects that the distance between the two vehicles is too large or too small. Under normal circumstances, the output is zero. Therefore, the actual operation k n Control acceleration command at the moment It can be calculated according to the following formula:

[0108]

[0109] Where, To actually run k n The control acceleration command at the moment, K P , K I , K D are the parameters of proportional, integral and differential links respectively, e(k n ) is the actual train speed v(k n ) and recommended driving curve speed v r (k n ), e(k n )=v(k n )-v r (k n );∫e(k n ) is the actual train position s(k n ) and recommended driving curve position s r (k n ), ∫e(k n )=s(k n )-s r (k n ); is the actual train acceleration a(k n ) and recommended driving curve acceleration a r (k n ), u adj Adjust the value of the control instruction for the spacing adjustment part.

[0110] At this point, the virtual train unit control layer follows the recommended driving curves generated by the planning layer, no longer relying on real-time tracking. Because the planning layer considers the synchronization and overall operation of the virtual train, the pilot train's operating strategy also takes the tracking of the follower train into account, ensuring that all train units work together for efficient and coordinated operation.

[0111] The layered control method for virtual marshaled trains can feasibly and efficiently plan and control the operational strategy for virtual marshaled trains, achieving the desired goal of synchronized operation. This invention can significantly improve the tracking efficiency of virtual marshaled trains, simplify the complexity of virtual marshaled train control problems, save real-time computing resources, enhance the synchronization of train unit operations, reduce the platform stop time difference (referred to as the stop time difference) between train units, and meet the requirements of actual operational indicators.

[0112] The present invention will be further described below in conjunction with a specific inter-station operation scenario of a virtual marshaled train.

[0113] Taking a virtual train consisting of two train units as an example, other multi-train hierarchical coordinated control systems and methods can be derived by analogy. The specific inter-station operation scenario uses data from an actual line and is implemented according to the implementation steps of the virtual train hierarchical coordinated control method described in the invention.

[0114] (1) Processing electronic map information according to the destination and other operation plans;

[0115] The virtual train schedule received from the automatic train operation monitoring system determines the operating intervals. The information for the intervals is shown in Table 1. Here, the planned interval is from platform 2 to platform 1 on the actual line. First, the initial positions of all logical links between the stations are found based on the electronic map. Based on the logical link numbers and offsets for track speed limits, slopes, and curve radii, the track parameters for each position between the stations are determined by adding the offsets to the initial logical link positions. Table 2 shows some of the data processing results.

[0116] Table 1 Actual line platform stop information

[0117]

[0118] Table 2 Logical section information of track line data

[0119]

[0120] The processed track parameters are then transmitted to the recommended driving curve generation module and the real-time control module. The speed limit information for the section determines the recommended driving curve's operational strategy. The track gradient and curve radius are used to calculate the emergency braking intervention speed to ensure train safety. Once the relevant information about the virtual train's route is collated, the recommended driving curve for the virtual train can be generated.

[0121] (2) Calculate the recommended driving curve of the virtual train at the planning level;

[0122] The planning layer constructs an optimization problem based on the inter-station operating indicators of virtual train formations and solves the problem to obtain the recommended driving curves for each train unit that meet the requirements. The inter-station operating indicator requirements are shown in Table 3.

[0123] Table 3 Inter-station operation index requirements

[0124] name Numerical Parking accuracy error requirements <30cm Runtime requirements 100±5s Maximum allowable departure time difference 2s Upper limit of parking time difference 4s

[0125] According to the index requirements and line data, the virtual train formation optimization problem can be constructed as follows:

[0126] stv i (k)≤v EBI (s i (k))

[0127] -0.8≤u i (k)≤0.8

[0128] -0.3≤u i (k+1)-u i (k)≤0.3

[0129] d(k)≤s1(k)-94.64-s2(k)

[0130] v i (k)≤V

[0131] i=1,2

[0132] Among them, the emergency braking intervention speed v of the following train EBI (s i (k)) The full-time and space security protection method is adopted. The present invention is applicable to various security protection calculation methods. Here, only one advanced method is selected as an example. The specific calculation parameters are shown in Table 4. Security tracking distance d i (k) needs to be determined based on the actual control error of the controller, generally maintaining a fixed difference from the minimum safe distance. Conventional mathematical optimization algorithms, such as heuristic algorithms, reinforcement learning algorithms, sequential quadratic programming, and active set methods, can be used to solve this optimization problem and obtain the recommended driving curve for the virtual train.

[0133] Table 4 Safety protection calculation parameters

[0134]

[0135] (3) Calculate the virtual train control actions at the control layer.

[0136] Based on the recommended driving curve for the virtual train formation generated by the planning layer, the control layer controls the trains to ensure that the actual speed of the train units is essentially the same as the recommended driving curve, and the control error is minimized. The control method can adopt the mainstream PID control method or model predictive control method in the railway industry. While the lead train and the following train follow their respective recommended driving curves, the controller monitors the operating status of both trains in real time and adjusts control actions as necessary to ensure that the two train units do not encounter dangerous situations due to the influence of their respective control errors.

[0137] The results of the reference curve planning and the double-real-car operation control experiment on the test line are as follows: Figure 5-Figure 8 shown. Figure 5 and Figure 6 The actual speed of the pilot train is basically consistent with the recommended driving speed of the pilot train, and the actual speed of the following train is basically consistent with the recommended driving speed of the following train.

[0138] The performance parameters of the generated recommended driving curve and the on-site dual-car tracking curve index data are shown in Table 5. The experimental data demonstrates the engineering feasibility of the present invention: a hierarchical collaborative control system and method for virtual marshaling trains, and its effectiveness in solving the current problem of synchronous operation of virtual marshaling trains.

[0139] Table 5 Recommended driving curve performance parameters

[0140]

[0141]

[0142] The present invention addresses the problem that the current virtual marshaling train operation control system and method do not take into account the operational integrity of all train units, and the system structure design does not match the virtual marshaling train operation requirements, resulting in asynchronous operation, high real-time control calculation pressure, and low operating efficiency. A layered collaborative control system and method for virtual marshaling trains are designed. The planning layer considers the synchronous operation target of the train units to generate a recommended driving curve for the virtual marshaling train, and the control layer achieves synchronous and efficient operation of the virtual marshaling train by tracking the recommended driving curve. The two-layer system structure of collaborative overall planning combined with curve tracking control ensures the synchronous and consistent operation of the virtual marshaling train units and transfers the real-time calculation pressure to the planning layer. The layered collaborative control system for virtual marshaling trains realizes a complete operation control method for virtual marshaling trains without changing the interfaces of other modules. The main innovations and beneficial effects of the present invention are summarized as follows:

[0143] 1. Improving the tracking efficiency of virtual marshaling trains: Based on the adopted safety protection distance characteristics, the present invention optimizes the driving curve of virtual marshaling trains and minimizes the tracking spacing between train units according to the control error margin, thereby improving the tracking efficiency of virtual marshaling trains.

[0144] 2. Simplify the complexity of virtual train control problems and save real-time computing resources: The present invention designs the real-time control target of the virtual train unit to track the recommended driving curve, thereby simplifying the nonlinear spacing tracking control problem related to the actual state of the preceding vehicle into a curve tracking problem, avoiding online calculation of nonlinear functions, greatly reducing the complexity of real-time control calculations, and achieving the purpose of saving computing resources.

[0145] 3. Improve the synchronization of train unit operation and meet the requirements of actual operation indicators: The present invention makes full use of the organized characteristics of rail transit, adjusts the driving strategy of virtual marshaling trains according to the known operation plan, fully ensures the synchronization of the operating status of virtual marshaling trains, and can effectively reduce the time difference of train platform stops, thereby meeting the requirements of actual operation indicators.

[0146] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0147] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A virtual train hierarchical cooperative control system, characterized in that: include: Information interaction module, electronic map processing module, recommended driving curve planning module and multiple real-time tracking control modules; One virtual marshaling train unit corresponds to one real-time tracking control module; The information interaction module is connected to the electronic map processing module; the information interaction module is used to obtain train operation plan information from the train automatic operation monitoring system and output the train operation plan information to the electronic map processing module; The electronic map processing module is connected to the recommended driving curve planning module and the multiple real-time tracking control modules respectively; the electronic map processing module is used to store track line information, and query the track line information of the entire train operation according to the train operation plan information, to form the track line operation data of the virtual marshaling train, and output it to the recommended driving curve planning module and the multiple real-time tracking control modules at the same time; The recommended driving curve planning module is connected to multiple real-time tracking control modules; the recommended driving curve planning module is used to regard the virtual marshaling train as a whole based on the track line operation data of the virtual marshaling train, coordinately plan the recommended driving curves of all train units in the virtual marshaling train, and send the recommended driving curves to the real-time tracking control module of each train unit; The real-time tracking control module is used to control the train units to run according to their respective recommended driving curves based on the track line operation data of the virtual marshaling train, so as to achieve the expected goal of synchronous operation of the virtual marshaling train; The recommended driving curve planning module considers the virtual train as a whole and collaboratively plans the recommended driving curves of all train units in the virtual train according to the track line operation data of the virtual train, specifically including: Based on the track line operation data between virtual marshaling train stations, a sequential quadratic programming method, an active set method, a heuristic algorithm, or a reinforcement learning algorithm is used to solve the mathematical model for the coordinated operation optimization between virtual marshaling train stations, and the control acceleration sequence of the entire virtual marshaling train operation plan between stations is obtained; the mathematical model for the coordinated operation optimization between virtual marshaling train stations is: s.t.v i (k)≤v EBI (s i (k)) in min in i (k)≤u max j min ≤u i (k+1)-u i (k)≤j max d(k)≤s1(k)-L-s2(k) v i (k)≤V i＝1,2 k=1,2,...,n-1,n Among them, Q1, Q2, Q3, and Q4 are the optimization weight coefficients of on-time parking, on-time operation, synchronous departure, and synchronous parking indicators respectively. and To plan and simulate parking locations for the pilot train and the following train, and The actual target stopping points for the lead train and the following train. and To simulate the stopping time of the pilot train and the following train, and is the initial departure time of the lead train and the following train, T g Establishing operating times between stations; represents the interval running time of the planning simulation; i = 1, 2 are the pilot train and the following train respectively; n represents the total running time of the virtual marshaling train between stations; u i (k),u i (k+1) represents the acceleration controlled by the train unit at time k and time k+1; U i represents the set of controlled accelerations of the train unit at each moment when it is running between stations; s1(k) and s2(k) represent the head positions of the lead train and the following train at time k, respectively; v i (k) represents the speed of the train unit at time k; u min and u max Indicates the train unit control acceleration limit value; j min and j max represents the limit value of the train unit impact rate; d(k) represents the minimum tracking distance between trains; L is the length of the train unit; V represents the speed limit of the track line, v EBI (s i (k)) represents the emergency braking intervention speed at time k; According to the control acceleration sequence, the train dynamics equation is used to obtain the position, velocity, and acceleration of the entire inter-station operation plan of the train unit, thereby determining the recommended driving curve between stations for each train unit in the virtual marshaling train; the train dynamics equation is: a i (k+1)=u i (k)+g i (s i (k),v i (k)); v i (k+1)=v i (k)+τa i (k); Among them, a i (k), a i (k+1) represents the total acceleration at time k and time k+1, v i (k), v i (k+1) represents the speed at time k and time k+1 respectively, s i (k), s i (k+1) represents the position at time k and time k+1 respectively, τ represents the time calculation step, g i (s i (k),v i (k)) represents the function of running resistance and the current position and speed of the train unit.

2. The virtual train hierarchical cooperative control system according to claim 1, characterized in that: The train operation plan information includes: train operation destination, expected operation time and section temporary speed limit; The track line information includes: track line logical section number, stop point, switch, transponder, slope and track static speed limit.

3. A method for hierarchical collaborative control of a virtual train, characterized in that: include: Establish a mathematical model for optimizing the coordinated operation of virtual trains between stations; Determine the track line operation data between virtual marshaling train stations based on the train station operation plan information; Solving the inter-station collaborative operation optimization mathematical model of the virtual marshaling train based on the track line operation data between the virtual marshaling train stations to obtain the recommended inter-station driving curve for each train unit in the virtual marshaling train; According to the real-time status of each train unit and the real-time status of the train units before and after each train unit, based on the track line operation data between the virtual marshaling train stations, the recommended driving curve between the stations of each train unit is tracked in real time, ultimately achieving the expected goal of synchronous operation of the virtual marshaling train; The mathematical model for optimizing the coordinated operation of virtual marshaling trains between stations is: s.t.v i (k)≤v EBI (s i (k)) in min in i (k)≤u max j min ≤u i (k+1)-u i (k)≤j max d(k)≤s1(k)-L-s2(k) v i (k)≤V i＝1,2 k=1,2,...,n-1,n Among them, Q1, Q2, Q3, and Q4 are the optimization weight coefficients of on-time parking, on-time operation, synchronous departure, and synchronous parking indicators respectively. and To plan and simulate parking locations for the pilot train and the following train, and The actual target stopping points for the lead train and the following train. and To simulate the stopping time of the pilot train and the following train, and is the initial departure time of the lead train and the following train, T g Establishing operating times between stations; represents the interval running time of the planning simulation; i = 1, 2 are the pilot train and the following train respectively; n represents the total running time of the virtual marshaling train between stations; u i (k),u i (k+1) represents the acceleration controlled by the train unit at time k and time k+1; U i represents the set of controlled accelerations of the train unit at each moment when it is running between stations; s1(k) and s2(k) represent the head positions of the lead train and the following train at time k, respectively; v i (k) represents the speed of the train unit at time k; u min and u max Indicates the train unit control acceleration limit value; j min and j max represents the limit value of the train unit impact rate; d(k) represents the minimum tracking distance between trains; L is the length of the train unit; V represents the speed limit of the track line, v EBI (s i (k)) represents the emergency braking intervention speed at time k; Based on the track line operation data between virtual marshaling train stations, solving the inter-station coordinated operation optimization mathematical model of the virtual marshaling train, and obtaining the recommended driving curve between stations for each train unit in the virtual marshaling train, specifically including: Based on the track line operation data between virtual marshaling train stations, a sequential quadratic programming method, an active set method, a heuristic algorithm or a reinforcement learning algorithm is used to solve the mathematical model for the coordinated operation optimization between the virtual marshaling train stations, and a control acceleration sequence for the entire virtual marshaling train operation plan between stations is obtained; According to the control acceleration sequence, the train dynamics equation is used to obtain the position, velocity, and acceleration of the entire inter-station operation plan of the train unit, thereby determining the recommended driving curve between stations for each train unit in the virtual marshaling train; the train dynamics equation is: a i (k+1)=u i (k)+g i (s i (k),v i (k)); v i (k+1)=v i (k)+τa i (k); Among them, a i (k), a i (k+1) represents the total acceleration at time k and time k+1, v i (k), v i (k+1) represents the speed at time k and time k+1 respectively, s i (k), s i (k+1) represents the position at time k and time k+1 respectively, τ represents the time calculation step, g i (s i (k),v i (k)) represents the function of running resistance and the current position and speed of the train unit.

4. The method for hierarchical coordinated control of a virtual train set according to claim 3, characterized in that: According to the train station operation plan information, the track line operation data between virtual marshaling train stations is determined, including: Obtaining electronic map information of the track line; the electronic map information includes a plurality of logical segments obtained by dividing the track line, and track line data for each logical segment; the track line data includes a speed limit start point, a speed limit end point, a slope start point, a slope end point, a curve radius start point, and a curve radius end point; All logical sections included in the operation plan stations and the track line data of each logical section are searched from the electronic map information as the track line operation data between the virtual marshaling train stations.

5. The layered cooperative control method for virtual train formation according to claim 3 is characterized in that: According to the real-time status of each train unit and the real-time status of the train units before and after each train unit, based on the track line operation data between the virtual marshaling train stations, the recommended driving curve between the stations of each train unit is tracked in real time, including: Based on the track line operation data between virtual marshaling train stations, a full-time and space safety protection method is used to determine the emergency braking intervention speed of the train unit at any moment; According to the real-time status of each train unit and the real-time status of the train units before and after each train unit, the feedforward PID control algorithm is used to track the recommended driving curve between stations of each train unit in real time, and the real-time tracking speed of the train unit is controlled to be less than or equal to the emergency braking intervention speed.

6. The layered cooperative control method for virtual train formation according to claim 5, characterized in that: The calculation formula of the feedforward PID control algorithm is: Where, To actually run k n The control acceleration command at the moment, K P , K I , K D are the parameters of proportional, integral and differential links respectively, e(k n ) is the actual train speed v(k n ) and recommended driving curve speed v r (k n ), e(k n )=v(k n )-v r (k n );∫e(k n ) is the actual train position s(k n ) and recommended driving curve position s r (k n ), ∫e(k n )=s(k n )-s r (k n ); is the actual train acceleration a(k n ) and recommended driving curve acceleration a r (k n ), u adj Adjust the control instruction value for the spacing adjustment part.

Images