FAO rail train control system based on rail transit

Abstract

The invention relates to the technical field of urban rail transit control, in particular to an FAO rail train control system based on rail transit, which is used for solving the problems that in the prior art, a reasonable train dynamic model cannot be constructed by integrating path, speed limit, acceleration, safety interval and timetable constraint conditions, an accurate train operation strategy cannot be generated, and the reliability of the system is poor. And the safety, the punctuality rate and the efficiency of train operation are reduced. An MPC model prediction control method is adopted through a train control decision module, a train dynamic model is constructed based on a track situation analysis result and a driving task target, a path, speed limit, acceleration, a safety interval and a timetable constraint condition are synthesized, an accurate train operation strategy is generated, rolling optimization and real-time control are achieved through the module, and the train control decision module has a good application prospect. And the safety, the punctuality rate and the efficiency of train operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail transit control, and more specifically, to an FAO rail train control system based on rail transit. Background Art

[0002] With the expansion of the scale of the urban rail transit network, the traditional CBTC system relying on manual driving leads to efficiency bottlenecks and safety risks. As the fourth-generation full-automatic operation technology, the FAO system can realize functions such as train dormancy wake-up, automatic operation, and fault self-healing by deeply integrating subsystems such as signals, vehicles, and integrated monitoring, significantly improving operation efficiency and safety.

[0003] The patent application with the publication number CN119636864A discloses an operation control system and method for an urban rail train, including: a primary operation control system and a standby operation control system; the standby operation control system includes a central train management server, an on-vehicle integrated control unit, an on-vehicle integrated BTM, an on-vehicle secondary radar antenna, and a trackside secondary radar base station; the central train management server is communicatively connected to the on-vehicle integrated control unit; the on-vehicle integrated control unit is communicatively connected to the CI, the on-vehicle integrated BTM, and the on-vehicle secondary radar antenna in the primary operation control system; the on-vehicle integrated control unit is used to perform authorization calculation according to the position information of the front and rear trains sent by the central train management server and the state information of the signal ahead obtained from the CI to achieve speed protection; it solves the problem in the prior art that there is no standby train operation control system after the CBTC system fails and speed protection cannot be continued. However, the above reference patent performs authorization calculation by combining the position information of the front and rear trains and the state of the signal ahead through the on-vehicle integrated control unit to achieve speed protection in the standby mode, solving the problems of no speed protection, easy delay, and accident occurrence during degraded operation after the failure of the primary system. However, it cannot accurately present the dynamic state of the train and track resources, cannot comprehensively consider multiple factors for dynamic planning of the optimal safe path, reducing the intelligent dispatching and operation efficiency in complex environments; at the same time, it cannot construct a reasonable train dynamics model by integrating path, speed limit, acceleration, safety interval, and timetable constraint conditions, and cannot generate accurate train operation strategies, reducing the safety, punctuality rate, and efficiency of train operation.

[0004] Therefore, we propose an FAO rail train control system based on rail transit for the above problems. Summary of the Invention

[0005] The object of the present invention is to provide an FAO rail train control system based on rail transit, which solves the problems that the prior art cannot accurately present the dynamic state of trains and track resources, cannot comprehensively consider multiple factors for dynamic programming of the optimal safe path, and reduces the intelligent dispatching and operation efficiency in complex environments; at the same time, it cannot construct a reasonable train dynamics model by integrating path, speed limit, acceleration, safety interval and timetable constraint conditions, cannot generate accurate train operation strategies, and reduces the safety, punctuality rate and efficiency of train operation.

[0006] The object of the present invention is achieved by the following technical solutions: An FAO rail train control system based on rail transit, which is applied to the FAO train management platform, includes: A train operation information acquisition module, which is used to collect train operation environment perception data. The train operation environment perception data includes wheel pulse count, transponder positioning information, track occupancy status, switch opening and closing status, speed limit section ahead and line gradient data, and performs preprocessing operations on the collected train operation environment perception data; A track situation analysis module, which dynamically constructs the track occupancy status based on the collected train operation environment perception data, and identifies the safe travel path of the train through graph modeling and adjacent graph search algorithms; A train control decision-making module, which generates a train operation strategy by using the MPC model predictive control method based on the track situation analysis result and the train operation task objectives. The train operation task objectives include destination, timetable, speed limit curve, stop plan and train interval requirements; An execution control module, which is used to receive the train operation strategy instruction and convert it into control actions of the traction and braking systems.

[0007] As a preferred embodiment of the present invention, the process of the track situation analysis module dynamically constructing the track occupancy status includes: Obtain the preprocessed train operation environment perception data. The train operation environment perception data includes wheel pulse count, transponder positioning information, track occupancy status, switch opening and closing status, speed limit section ahead and line gradient data; The specific steps of dynamically constructing the track occupancy status are as follows: S1: Divide the whole track into discrete track segment units, each track segment has a unique number, and establish a mapping relationship with the transponder and switch facilities; S2: Use the transponder positioning point for anchor calibration to determine the track segment index where the train is located, and combine the wheel pulse count and the position of the previous anchor point to calculate the track segment range covered by the head and tail of the train; S3: Mark the track segment currently occupied by the train as "occupied", and mark the track segments ahead of the train's operation as "restricted" or "unknown" due to switch configuration restrictions or already occupied by other trains; S4: The system updates the train position information at a fixed period. When it detects that the train crosses the track section boundary, it updates the track occupancy status table, collects the running status of other trains on the whole line, and generates a global track occupancy situation map.

[0008] As a preferred embodiment of the present invention, the process of the track situation analysis module identifying the safe travel path of the train through graph modeling and adjacent graph search algorithm includes: Construct a track network graph G=(V, E), where V is the node, each track section is regarded as a node, and E is the edge. An edge is established between two passable track sections; The attributes of the track node include: occupancy status, current speed limit, slope value, and the number and status of the turnout it belongs to; The attributes of the track edge include: traffic direction, turnout switching cost, and the minimum safe interval time from the train ahead; Use the A-Star algorithm to identify the safe travel path of the train. The specific steps are as follows: T1: During the adjacent matrix modeling process, a connection edge is established between the track section node and its adjacent nodes. For the node containing a turnout, multiple branch edges are generated according to the current opening and closing state of the turnout; T2: Design the heuristic function h(n) as the estimated distance from the current node to the target node; T3: During the path search process, the initial node is set as the current position of the train, and the target node is the target section. By traversing all reachable paths, the path with the minimum f(n)=g(n)+h(n) is preferentially selected, where g(n) represents the actual cost of the traveled path. When the path contains an occupied track section or the turnout is in an illegal state, it is marked as an invalid path and traced back to the nearest valid branch to continue the search; T4: Output the path and mark the status of each track section, and mark the searched path as "feasible path".

[0009] As a preferred embodiment of the present invention, the process of the train control decision module constructing a train dynamics prediction model includes: Obtain the train operation task objectives from the FAO system, and the train operation task objectives include destination, timetable, speed limit curve, stop plan, and train interval requirements; Construct a train dynamics prediction model. The motion state of the train is described by the following variables: Position x(t): The current position of the train; Speed v(t): The current speed of the train; Acceleration a(t): The acceleration and deceleration control quantity of the train; The state transition relationship of the train motion is expressed by the following formula: Position update: ; Speed update: , where ∆t is the control period, which represents the time interval between each calculation.

[0010] As a preferred embodiment of the present invention, the process of the train control decision module constructing a model to set track situation analysis result constraints and speed limit constraints includes: Track situation analysis result constraints: Based on the track situation analysis results, the train is subject to the following constraints when traveling on the planned path: the train can only travel on the identified feasible path, and the train can only travel along the path consistent with the direction of the switch opening; Speed and acceleration restrictions: the train's operating speed must be within the maximum speed range allowed by the speed limit curve, that is, 0≤v(t)≤v max (x), where v max (x) is the speed limit given based on the track conditions, and the magnitude of the acceleration should satisfy a min ≤a(t)≤a max .

[0011] As a preferred embodiment of the present invention, the process of constructing a model for setting vehicle spacing constraints and timetable constraints by the train control decision module includes: Interval constraint: Interval refers to the minimum safe distance or time interval that should be maintained between a train and the train in front or behind it. It can be achieved in the following two ways: Location-based safety distance constraints: The distance between the current train position x(t) and the preceding train position should not be less than the set safety distance d safe ,Right now: Front vehicle position - x(t) ≥ d safe , indicating that the current train is not allowed to enter the area where the distance to the preceding train is less than the safe distance; Run-time-based safety interval constraints: The time interval between the current train and the preceding train should not be less than the minimum allowed value T min ,Right now: ∆t≥T min , indicating that the distance between the two trains must be at least T min The running time interval; Train operations must meet schedule constraints, meaning their arrival and departure times must be consistent with the scheduled schedule. However, due to road conditions and traffic signal factors, trains may be delayed or advanced. To minimize this deviation, a penalty term can be set by optimizing the objective function to ensure that the train's arrival or departure times are close to the schedule requirements: ; Among them, w1, w2, and w3 are weight factors. is the actual speed of the train at the kth station, is the target speed that the train expects to reach at the k-th station, is the actual arrival time of the train at the k-th station, is the scheduled arrival time in the train schedule.

[0012] As a preferred embodiment of the present invention, the process in which the train control decision-making module uses the MPC model predictive control method to generate a train operation strategy includes: The MPC operation process is as follows: Initialization: Obtain the current state of the train; Track input: Load the track situation analysis result to determine the train's feasible path and obstacle information; Task objective input: Load the train operation task objectives, including destination, train schedule, speed limit curve, stop plan, and train spacing requirements; Optimization and solution: Use the MPC algorithm to calculate the optimal control input for a future period of time; Execute control: Execute the acceleration command at the current moment and continue rolling optimization based on the state of the new cycle; Based on the MPC control instruction sequence generated by rolling optimization, a dynamic operation strategy of the train within the current operation cycle is formed. The train operation strategy includes timestamp, running speed curve, acceleration control, stop control point, start or brake timing, safety interval control, arrival time plan, line information, and train status information.

[0013] As a preferred embodiment of the present invention, the process in which the execution control module receives the train operation strategy instruction and converts it into control actions of the traction and braking systems includes: Obtain the train operation strategy instruction, which includes timestamp, speed instruction, acceleration instruction, stop control point, start or brake timing, safety interval control instruction, arrival time plan, line information, and train status information; Convert the train operation strategy instruction into control actions of the traction and braking systems. The specific conversion process is as follows: Speed control: The execution control module determines the required traction or braking force based on the desired speed curve, current speed, and line conditions; Acceleration control: Adjust the change rate of traction and braking force according to the set acceleration instruction, actual acceleration state, and system limit conditions; Stop control: Adjust the change of braking force according to the set stop position, train status, and line conditions; Start or brake timing control: Determine the trigger time of the traction or braking action according to the operation plan, train status, and trajectory prediction; Safety interval control: Adjust the train running speed according to the set safety interval instruction, real-time train status, and information of the leading train; Arrival time control: Adjust the train running speed according to the set arrival time, real-time train status, and line conditions.

[0014] Compared with the prior art, the advantages of the present invention are as follows: (1) In the present invention, the track situation analysis module fuses multi-source perception data to construct a high-precision track occupancy status map, accurately presenting the dynamics of the train and track resources. Using graph modeling and the A-Star algorithm, it comprehensively considers multiple factors for dynamic programming of the optimal safe path, avoiding conflict risks. Its output provides support for train decision-making, significantly improving the intelligent dispatching and operation efficiency in complex environments; (2) In the present invention, the train control decision module adopts the MPC model predictive control method. Based on the track situation analysis results and train operation task objectives, it constructs a train dynamics model, comprehensively considering path, speed limit, acceleration, safety interval, and timetable constraint conditions, generating an accurate train operation strategy. The module realizes rolling optimization and real-time control, improving the safety, punctuality rate, and efficiency of train operation, and outputting complete control instructions to provide reliable support for execution control. It is the core link for the full-automatic operation system to achieve intelligent decision-making. Description of the Drawings

[0015] Figure 1 It is the system block diagram in the present invention; Figure 2 It is the step flow chart for dynamically constructing the track occupancy status in the present invention; Figure 3 It is the step flow chart for using the A-Star algorithm to identify the safe travel path of the train in the present invention. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention; obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment 1: As Figure 1 、 Figure 2 and Figure 3 shown, a FAO track train control system based on rail transit proposed by the present invention is applied to the FAO train management platform, including: The train operation information collection module is used to collect train operation environment perception data. The train operation environment perception data includes wheel pulse count, balise positioning information, track occupancy status, switch opening / closing status, speed limit section ahead, and line gradient data. It performs preprocessing operations on the collected train operation environment perception data. The preprocessing operations include data cleaning, data calibration, data fusion, and format conversion; Through the collection and fusion of multi-source data (such as wheel pulses, balise positioning, track status, speed limit gradient, etc.), the train operation information collection module realizes the comprehensive perception of the train operation environment. Combined with preprocessing operations such as data cleaning, calibration, fusion, and format conversion, it improves the data accuracy and system reliability. This module provides high-quality input for train positioning, speed control, path planning, and safety protection, effectively ensuring operation safety and optimizing operation efficiency.

[0018] The track situation analysis module dynamically constructs the track occupancy status based on the collected train operation environment perception data, and identifies the safe travel path of the train through graph modeling and adjacent graph search algorithms; The process of the track situation analysis module dynamically constructing the track occupancy status includes: Obtain the preprocessed train operation environment perception data. The train operation environment perception data includes wheel pulse count, balise positioning information, track occupancy status, switch opening / closing status, speed limit section ahead, and line gradient data; The specific steps of dynamically constructing the track occupancy status are as follows: S1: Divide the entire track into discrete track segment units. For example, there is a node every 50 meters. Each track segment has a unique number and is mapped to balises and switch facilities; S2: Use the balise positioning point for anchor calibration to determine the track segment index where the train is located, and combine the wheel pulse count with the position of the previous anchor point to calculate the track segment ranges covered by the head and tail of the train; S3: Mark the track segment currently occupied by the train as "occupied", and mark the track segments ahead of the train's operation that are restricted by switch configurations or already occupied by other trains as "restricted" or "unknown"; S4: The system updates the train position information at a fixed cycle (such as 100 milliseconds). When it detects that the train crosses the track segment boundary, it updates the track segment occupancy status table and collects the operation status of other trains on the entire line to generate a global track occupancy situation map; The process of the track situation analysis module identifying the safe travel path of the train through graph modeling and adjacent graph search algorithms includes: Construct a track network graph G=(V, E), where V is the node, and each track interval (such as: 50m segment) is regarded as a node, and E is the edge. An edge is established between two passable track segments (connected front and back or through a switch); The attributes of track nodes include: occupancy status, current speed limit (e.g., 80 km / h), gradient value, and the turnout number and status to which it belongs (applicable to branch nodes); The attributes of the track side include: traffic direction (one-way or two-way), turnout switching cost (e.g., increasing by 1 time unit), and the minimum safe interval time from the train ahead (affecting path validity); Use the A-Star algorithm to identify the safe travel path of the train. The specific steps are as follows: T1: During the adjacency matrix modeling process, a connection edge is established between the track segment node and its adjacent nodes. For nodes with turnouts, multiple branch edges are generated according to the current opening and closing state of the turnout. The edge weight is dynamically set according to the track segment attributes, and higher weight values are assigned to speed limit sections and uphill sections; T2: Design the heuristic function h(n) as the estimated distance from the current node to the target node (usually using the straight-line distance as an approximation). To optimize path selection, a penalty term can be introduced when calculating h(n) to increase the additional cost for occupied track segments or paths involving turnout operations, so as to guide the search process to avoid these paths; T3: During the path search process, the initial node is set as the current position of the train, and the target node is the target section. By traversing all reachable paths, the path with the minimum f(n)=g(n)+h(n) is preferentially selected, where g(n) represents the actual cost of the traveled path. When the path contains an occupied track segment or the turnout is in an illegal state, it is marked as an invalid path and traced back to the nearest valid branch to continue the search; T4: Output the path and mark the status of each track segment. Mark the searched path as a "feasible path", and then provide it to the decision-making module as input data for tasks such as speed planning and station braking decision-making; The track situation analysis module constructs a high-precision track occupancy status map in real time by fusing multi-source perception data (such as wheel pulses, transponder positioning, turnout status, etc.), accurately reflecting the train position and the use of track resources. At the same time, this module uses graph modeling and the A-Star search algorithm to dynamically calculate the optimal safe path by comprehensively considering factors such as speed limit, gradient, and turnout status, effectively avoiding conflict and illegal operation sections, ensuring the safety and efficiency of path planning, and its output provides key support for train control decisions, improving the system's intelligent scheduling ability and overall operation efficiency in complex operating environments.

[0019] The train control decision-making module, based on the track situation analysis results and the train operation task objectives, which include destination, timetable, speed limit curve, stop plan, and train spacing requirements, uses the MPC model predictive control method to generate train operation strategies; The process of the train control decision-making module constructing a train dynamics prediction model includes: Obtain train operation task objectives from the FAO system. The train operation task objectives include the destination, timetable, speed limit curve, stop plan, and headway requirements. The destination is the final destination of the train, usually the target station, which indicates the final stopping position of the train in this area. The timetable is the train's timetable, including the arrival and departure times at each station. These time data sometimes need to be dynamically adjusted according to real-time situations (such as delays) to ensure that the train runs according to the plan. The speed limit curve is the speed limit for each section of the track, including the limitations on speed due to factors such as curve radius and gradient. The speed limit curve ensures that the train travels under the premise of safety. The stop plan is the stations where the train stops and the dwell time at each station, including whether to stop at a station and the exact position of parking. The headway requirements are the minimum time interval or distance that the train needs to maintain from the front and rear trains, which ensures that the train maintains a safe distance during operation and avoids collisions. These task objectives are continuously updated through wireless communication or train-ground communication links to ensure the real-time and accuracy of train dispatching instructions. The above train operation task objectives are periodically updated through the vehicle-ground communication interface to ensure the real-time and accuracy of dispatching instructions; MPC is a rolling optimization control strategy. At each control moment, it predicts the train operation state in the future for a period of time and optimizes the control input considering the constraint conditions; Build a train dynamics prediction model. The motion state of the train is described by the following variables: Position x(t): The current position of the train; Speed v(t): The current speed of the train; Acceleration a(t): The acceleration and deceleration control quantity of the train (regarded as the control input quantity); The state transition relationship of the train motion is expressed by the following formula: Position update: ; Speed update: , where ∆t is the control period, representing the time interval for each calculation; The process of the train control decision-making module building the model to set the track situation analysis result constraint and speed limit constraint includes: Track situation analysis result constraint. According to the track situation analysis result, when the train travels on the planned path, it is subject to the following constraints: The train can only travel on the identified feasible path, select the current optimal path from the output set of feasible paths, avoid entering the occupied track section, and the train can only travel along the path consistent with the turnout opening direction; Limitations on speed and acceleration. The running speed of the train must be within the maximum speed range allowed by the speed limit curve, that is, 0 ≤ v(t) ≤ v max (x), where v max(x) is the speed limit given based on the track conditions. The magnitude of acceleration (including positive acceleration and negative acceleration) should meet a min ≤a(t)≤a max To ensure that the train does not cause unsafe or unstable operation due to excessive acceleration or deceleration; The process of building a model for the train control decision module to set vehicle spacing constraints and timetable constraints includes: Interval constraints refer to the minimum safe distance or time interval that a train should maintain between itself and the train in front or behind it. They are used to ensure train safety during operation and prevent accidents such as rear-end collisions. They can be implemented in the following two ways: Location-based safety distance constraints: The distance between the current train position x(t) and the preceding train position should not be less than the set safety distance d safe ,Right now: Front vehicle position - x(t) ≥ d safe , indicating that the current train is not allowed to enter the area where the distance to the preceding train is less than the safe distance; Run-time-based safety interval constraints: The time interval between the current train and the preceding train should not be less than the minimum allowed value T min ,Right now: ∆t≥T min , indicating that the distance between the two trains must be at least T min The operating time interval is set to ensure the safety of operation; Train operations must meet schedule constraints, meaning their arrival and departure times must be consistent with the scheduled schedule. However, due to road conditions and traffic signal factors, trains may be delayed or advanced. To minimize this deviation, a penalty term can be set by optimizing the objective function to ensure that the train's arrival or departure times are close to the schedule requirements: ; Among them, w1, w2, and w3 are weight factors. is the actual speed of the train at the kth station, is the target speed that the train expects to reach at the kth station, is the actual arrival time of the train at the kth station, is the scheduled arrival time in the timetable; The process of the train control decision module using the MPC model predictive control method to generate a train operation strategy includes: The MPC operation process is as follows: Initialization: Get the current state of the train (position, speed, acceleration, etc.); Track input: Load track situation analysis results to determine the train's feasible path and obstacle information; Task objective input: Load the train operation task objectives, including destination, timetable, speed limit curve, stop plan, and headway requirements; Optimization and solution: Use the MPC algorithm (such as quadratic programming QP or mixed-integer quadratic programming MIQP) to calculate the optimal control input (acceleration) for a period of time in the future; Execution of control: Execute the acceleration command at the current moment, and continue the rolling optimization based on the state of the new cycle. Only the first control action is executed for each optimization, and then re-prediction and optimization are performed based on the new observed data to form a closed-loop feedback; The MPC control instruction sequence generated based on rolling optimization constitutes the dynamic operation strategy of the train in the current operation cycle. The train operation strategy includes timestamp, running speed curve, acceleration control, stop control points, start or braking timing, safety interval control, arrival time plan, line information, and train status information; The train control decision-making module adopts the MPC model predictive control method, integrates the results of track situation analysis and train operation task objectives, and has the capabilities of multi-constraint modeling and rolling optimization; by constructing a train dynamics model and integrating constraints such as path, speed limit, acceleration, safety interval, and timetable, it realizes the dynamic generation and real-time adjustment of the train operation strategy; this module effectively improves operation safety, punctuality, and efficiency, and outputs complete control instructions, providing reliable strategy support for the execution layer, which is the core link for the full-automatic operation system to achieve intelligent decision-making.

[0020] Embodiment 2: The technical solution of this embodiment of the present invention is different from that of Embodiment 1 in that: As Figure 1 shown, the execution control module is used to receive the train operation strategy instruction and convert it into control actions of the traction and braking systems; The process of the execution control module receiving the train operation strategy instruction and converting it into control actions of the traction and braking systems includes: Obtain train operation strategy instructions. Train operation strategy instructions are usually transmitted through in-vehicle networks, such as Ethernet or CAN bus. The train operation strategy instructions include a timestamp (indicating the time when the instruction is issued, used for synchronization and time management), a speed instruction (the desired speed curve, usually given in the form of a time series, such as speed values at certain time intervals, and may also include speed limit values), an acceleration instruction (the desired acceleration value, which may also be given in a time series or other way and includes acceleration limit values), a parking control point (including the exact position coordinates of the parking point and the desired parking speed, and may include deceleration curve parameters near the parking point), the start or braking timing (indicating when to start accelerating or decelerating, usually referenced by distance or time, and may also include acceleration curve parameters during the start and braking processes), a safety interval control instruction (indicating the safety distance or time interval to maintain with the preceding or following train, and may include specific safety distance calculation methods and parameters), the arrival time plan (the moments when the train is expected to arrive at each station, and may include arrival time windows and corresponding fault tolerance mechanisms), track information (including information such as the gradient and curve radius of the current track, used for the calculation of the auxiliary control algorithm) and train status information (information such as the current speed, position, acceleration, and train weight of the train, used for the feedback of the control algorithm); Convert the train operation strategy instructions into control actions for the traction and braking systems. The specific conversion process is as follows: Speed control: The execution control module determines the required traction or braking force based on the desired speed curve, the current speed, and track conditions (such as gradient and curve radius). Commonly used control methods include PID control, feedforward control, and model predictive control (MPC) for speed trajectory tracking; Acceleration control: Adjust the change rate of the traction and braking forces according to the set acceleration instruction, the actual acceleration state, and system limit conditions. By establishing a response model of the traction and braking systems and combining control algorithms, dynamic tracking of the desired acceleration trajectory is completed; Parking control: Adjust the change of the braking force according to the set parking position, train status, and track conditions. Commonly used control methods include S-curve deceleration control. Combining the inertia and braking characteristics of the train, trajectory tracking of the target parking point is completed; Start or braking timing control: Determine the trigger time of the traction or braking action according to the operation plan, train status, and trajectory prediction. By predicting the future operation state of the train, early intervention in the traction and braking systems is achieved; Safety interval control: Adjust the train operation speed according to the set safety interval instruction, the real-time status of the train, and the information of the preceding train. This process relies on the interaction with the train automatic protection system and, based on the train speed, relative position, and the operation state of the preceding train, realizes dynamic control of the train spacing; Arrival Time Control: Adjust the train running speed according to the set arrival time, real-time train status, and line conditions. The control logic needs to be based on disturbance factors such as passenger boarding and alighting, signal delay, and emergencies for dynamic scheduling and speed planning; The underlying execution control of the train operation strategy is realized by the following technologies: Real-time operating system: Responsible for the scheduling and execution of control tasks; Embedded system: Carries control algorithms and realizes communication with hardware modules; Sensor fusion: Integrates speed, position, and acceleration sensor data for estimating the train status; Control algorithm: Adopts PID and MPC methods to generate traction and braking force output commands; Hardware interface: Connects traction motors, braking devices, and sensors to complete the input and output of control signals; The execution control module has a high degree of integration and can uniformly process multi-dimensional operation strategy instructions. Combining control algorithms such as PID and MPC, it realizes precise control of traction and braking forces. The module can dynamically adjust the control strategy according to the train status and line conditions, improving operation safety, comfort, and energy efficiency, and providing stable and reliable underlying control support for the full-automatic operation system.

[0021] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its improved concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A FAO train control system based on rail transit, which is applied to the FAO train management platform, is characterized in that, Including: A train operation information acquisition module, which is used to acquire train operation environment perception data. The train operation environment perception data includes wheel pulse count, balise positioning information, track occupancy status, switch opening and closing status, speed limit section ahead and line gradient data, and performs preprocessing operations on the acquired train operation environment perception data; A track situation analysis module, which dynamically constructs a track occupancy status based on the acquired train operation environment perception data, and identifies the safe travel path of the train through graph modeling and adjacent graph search algorithms; A train control decision-making module, which generates a train operation strategy by using the MPC model predictive control method based on the track situation analysis result and the train operation task objectives. The train operation task objectives include destination, timetable, speed limit curve, stop plan and train spacing requirements; An execution control module, which is used to receive the train operation strategy instruction and convert it into control actions of the traction and braking systems.

2. The FAO rail train control system based on rail transit according to claim 1, wherein The process by which the track situation analysis module dynamically constructs the track occupancy status includes: Obtaining the preprocessed train operation environment perception data, which includes wheel pulse count, balise positioning information, track occupancy status, switch opening and closing status, speed limit section ahead and line gradient data; The specific steps for dynamically constructing the track occupancy status are as follows: S1: Divide the entire track into discrete track segment units, each track segment has a unique number, and establish a mapping relationship with balises and switch facilities; S2: Use the balise positioning point for anchor calibration to determine the track segment index where the train is located, and combine the wheel pulse count and the position of the previous anchor point to calculate the track segment ranges covered by the head and tail of the train; S3: Mark the track segment currently occupied by the train as "occupied", and mark the track segments ahead of the train's operation as "restricted" or "unknown" due to switch configuration restrictions or being occupied by other trains; S4: The system updates the train position information at a fixed period. When it detects that the train crosses the track segment boundary, it updates the track segment occupancy status table, and collects the operation status of other trains on the whole line to generate a global track occupancy situation map.

3. The FAO rail train control system based on rail transit according to claim 2, wherein, The process by which the track situation analysis module identifies the safe travel path of the train through graph modeling and adjacent graph search algorithms includes: Construct a track network graph G=(V, E), where V is the node, each track interval is regarded as a node, and E is the edge, and an edge is established between two passable track segments; The attributes of the track nodes include: occupancy status, current speed limit, gradient value, and the switch number and status to which they belong; The attributes of the track edges include: traffic direction, switch switching cost, and the minimum safe interval time with the train ahead; Use the A-Star algorithm to identify the safe travel path of the train. The specific steps are as follows: T1: During the adjacent matrix modeling process, connect edges are established between the track segment nodes and their adjacent nodes. For nodes containing switches, multiple branch edges are generated according to the current opening and closing status of the switches; T2: Design the heuristic function h(n) as the estimated distance from the current node to the target node; T3: During the path search process, the initial node is set as the current position of the train, and the target node is the target section. By traversing all reachable paths, the path with the minimum f(n)=g(n)+h(n) is preferentially selected, where g(n) represents the actual cost of the traveled path. When the path contains an occupied track section or the switch is in an illegal state, it is marked as an invalid path, and the search continues by backtracking to the nearest valid branch. T4: Output the path and mark the status of each track section, and mark the searched path as "feasible path".

4. A FAO rail train control system based on rail transit according to claim 1, characterized in that, The process by which the train control decision-making module constructs a train dynamics prediction model includes: Obtain the train operation task objectives from the FAO system, where the train operation task objectives include destination, timetable, speed limit curve, stop plan, and train separation requirements; Construct a train dynamics prediction model. The motion state of the train is described by the following variables: Position x(t): The current position of the train; Speed v(t): The current speed of the train; Acceleration a(t): The acceleration and deceleration control quantity of the train; The state transition relationship of the train motion is expressed by the following formula: Location update: ; Speed update: , where ∆t is the control period, representing the time interval for each calculation.

5. The FAO rail train control system based on rail transit according to claim 4, characterized in that, The process by which the train control decision-making module constructs a model to set the track situation analysis result constraint and speed limit constraint includes: Track situation analysis result constraint. According to the track situation analysis result, when the train travels on the planned path, it is subject to the following constraints: The train can only travel on the identified feasible path, and the train can only travel along the path consistent with the switch opening direction; Regarding the speed and acceleration limits, the operating speed of the train must be within the maximum speed range allowed by the speed limit curve, i.e., 0 ≤ v(t) ≤ v max (x), where v max (x) is the speed limit given based on the track conditions, and the magnitude of the acceleration should satisfy a min ≤ a(t) ≤ a max .

6. The FAO rail train control system based on rail transit according to claim 5, characterized in that, The process by which the train control decision-making module constructs a model to set the train separation constraint and timetable constraint includes: Train separation constraint. The train separation refers to the minimum safety distance or time interval that should be maintained between the train and the train in front or behind it, which can be achieved in the following two ways: Safety distance constraint based on position: The distance between the current train position x(t) and the position of the train ahead should be no less than the set safety distance d safe , that is: The position of the leading vehicle - x(t) ≥ d safe , indicating that the current train is not allowed to enter the area where the distance from the leading vehicle is less than the safe distance; Safety interval constraint based on running time: The time interval between the current train and the preceding train should be no less than the minimum allowable value T min , that is: ∆t≥T min indicates that the running time interval between two trains must be at least T min ; The train operation needs to meet the timetable constraint, that is, its arrival and departure times should be consistent with the predetermined timetable. However, due to road conditions and traffic signal factors, delays or advances may occur. To minimize this deviation, a penalty term can be set by optimizing the objective function so that the arrival or departure time of the train is close to the timetable requirements: ； where w1, w2, and w3 are weighting factors, is the actual speed of the train at the k-th station, is the target speed that the train expects to reach at the k-th station, is the actual arrival time of the train at the k-th station, is the scheduled arrival time in the timetable.

7. A FAO rail train control system based on rail transit according to claim 6, characterized in that, [[ID=,16]]The process by which the train control decision-making module uses the MPC model predictive control method to generate a train operation strategy includes: The MPC operation process is as follows: Initialization: Obtain the current state of the train; Track input: Load the track situation analysis result to determine the feasible path and obstacle information of the train; Task objective input: Load the train operation task objectives, including destination, timetable, speed limit curve, stop plan, and train separation requirements; Optimization solution: Use the MPC algorithm to calculate the optimal control input for a future period of time; Execute control: Execute the acceleration command at the current moment and continue rolling optimization according to the state of the new cycle; Based on the MPC control instruction sequence generated by rolling optimization, it constitutes the dynamic operation strategy of the train in the current operation cycle. The train operation strategy includes timestamp, running speed curve, acceleration control, stop control point, start or brake timing, safety interval control, arrival time plan, line information, and train state information.

8. The FAO rail train control system based on rail transit according to claim 1, wherein, The process in which the execution control module receives the train operation strategy instruction and converts it into the control actions of the traction and braking systems includes: Obtaining the train operation strategy instruction, which includes the timestamp, speed instruction, acceleration instruction, stop control point, start or braking timing, safety interval control instruction, arrival time plan, line information, and train status information; Converting the train operation strategy instruction into the control actions of the traction and braking systems. The specific conversion process is as follows: Speed control: The execution control module determines the required traction or braking force based on the desired speed curve, current speed, and line conditions; Acceleration control: Adjust the change rate of the traction and braking forces according to the set acceleration instruction, actual acceleration state, and system limit conditions; Stop control: Adjust the change of the braking force according to the set stop position, train status, and line conditions; Start or braking timing control: Determine the trigger time of the traction or braking action according to the operation plan, train status, and trajectory prediction; Safety interval control: Adjust the train running speed according to the set safety interval instruction, real-time train status, and information of the preceding train; Arrival time control: Adjust the train running speed according to the set arrival time, real-time train status, and line conditions.

Images