Heavy-load combined train cooperative communication consistency braking control method and device

Abstract

The application discloses a heavy-load combined train cooperative communication consistency braking control method and device. In the train braking process, the braking controller combines the strict constraints of the actuator and the relative displacement between the carriages, and adopts a distributed model predictive control method to design a cooperative braking controller according to the state variables such as the speed, acceleration and position information of each sub-node; the train consistency braking force control method is designed to solve the performance index of minimizing the speed inconsistency and the relative displacement difference between the carriages. The application reduces the relative displacement and the speed inconsistency between the carriages during train braking, reduces the size of the train hook force during braking, effectively reduces the damage of the carriage structure, and improves the comfort during train driving.

Description

Technical Field

[0001] This invention belongs to the field of train driving and braking control, and specifically relates to a method and device for coordinated communication and consistent braking control of heavy-load combined trains. Background Technology

[0002] Currently, train control strategies are primarily designed based on a single-mass motion model, which assumes the heavy-haul combined train is a rigid body and neglects the elastic dynamics between adjacent carriages. In reality, the carriages of heavy-haul combined trains are connected in a non-rigid manner. When relative displacement exists between adjacent carriages, coupled coupler forces are generated. Excessive coupler forces can damage the carriage structure and affect the ride comfort of the heavy-haul combined train. Therefore, in the design of braking control strategies for heavy-haul combined trains, it is essential to consider the nonlinear coupling between adjacent carriages to ensure that each carriage can achieve consistent braking. Research on the consistent braking control problem of trains has significant theoretical and practical implications.

[0003] In existing technologies, most distributed cooperative control strategies in train control employ linear dynamic models and linear controllers, failing to consider inherent actuator saturation, state constraints, and nonlinearities and uncertainties inherent in train control systems. Distributed model predictive control (DMDC) can handle cooperative control objectives, but it has not adequately studied the consistency control problem of multi-agent systems with nonlinear and uncertain characteristics. In practical applications, heavy-haul combined trains frequently operate in tunnels and on sloping roads, encountering not only wind resistance but also unknown and unmeasurable external disturbances. To ensure the performance control of train controllers using DMDC, a high-precision train model is required. However, current modeling methods, in order to simplify controller design, locally linearize the nonlinear train dynamics model, failing to accurately capture the dynamic characteristics of the train. Summary of the Invention

[0004] To overcome the technical problems in current heavy-haul combined train control strategies, such as insufficient estimation of train dynamic characteristics leading to inconsistency in braking control, resulting in poor train comfort and potential damage to carriage structures, this invention provides a cooperative communication-based consistent braking control method and device for heavy-haul combined trains. This invention constructs the relationship between the forces, speed, and acceleration of each carriage by combining the force and motion information of each carriage, and builds a Franklin linearized model for nonlinear systems. It constructs a consistent braking objective for the train's distributed braking system, obtains state variables such as speed, acceleration, and position information of child nodes by combining node communication topology, and designs a cooperative braking controller using a distributed model predictive control method; thus obtaining the optimal train distributed braking strategy.

[0005] To achieve the above-mentioned technical objectives, the technical solution of the present invention is as follows:

[0006] A method for coordinated communication and consistent braking control of heavy-haul combined trains includes the following steps.

[0007] A: Obtain the status information of each carriage of the heavy-haul combined train;

[0008] B: Treat each carriage as an individual point mass, and based on the state information and other forces acting on the train during operation, describe the influence of each force on the carriage's velocity and acceleration using the Franklin operator.

[0009] C: Determine the consistency braking target of the train's distributed braking system based on the speed difference and relative displacement between each carriage;

[0010] D: Based on the influence of each force on the carriage in step B and the braking target in step C, design a distributed Franklin model predictive control controller to distribute the braking force of each carriage.

[0011] In the method described above, step A includes the carriage's state information, which includes speed, acceleration, mass, braking force, friction, and coupling force between carriages. The connection between adjacent carriages is considered as an elastic mechanical connection. The coupling force generated by the relative displacement between carriages is called the coupler force. Friction is the rolling mechanical resistance between the carriage wheels and the track.

[0012] The method, wherein step B includes:

[0013] Based on the collected state information, and further considering the additional resistance caused by aerodynamic drag and changing road conditions, the relationship between various forces and the motion state of the carriage is expressed as follows:

[0014]

[0015] in , , These represent the mass, speed, and geographical location of the i-th carriage, where i = {1, 2, ..., n}. express The derivative of , where t represents time. Let the braking force of the i-th carriage be... Let i be the magnitude of the coupling force between the i-th car and the (i+1)-th car. The derivative of v is the acceleration. Let be the frictional force acting on the i-th carriage. The i-th car is subjected to additional resistance caused by external factors, including track incline, curvature, and tunnel. Indicates aerodynamic drag;

[0016] The relationship between the force and the motion state of the carriage is discretized as follows:

[0017]

[0018] in, The chi-square represents the state variable, and k represents the current time. Indicates the sampling time interval. , , and These are all intermediate variables used in the calculation, and the calculation method is as follows:

[0019]

[0020] Where f i This is the state transition function.

[0021] In the method described above, step C, the goal of consistent braking is to keep the speed difference and relative displacement of adjacent carriages in a small neighborhood where they are as close to zero as possible during the process of the travel speed of each heavy-load combined train carriage node converging to the desired speed curve.

[0022] The method, wherein step D includes:

[0023] D1: Using the Franklin model to predict the future dynamic behavior of each car in a heavily loaded combined train (as a child node), the prediction time domain is set as follows: Control time domain is ;

[0024] D2: Based on the prediction time domain set in D1 Control time domain For prediction in the time domain Define the predicted output trajectory, optimal output trajectory, and assumed output trajectory of the child nodes for the control time domain. Define the predictive control input, optimal control input, and hypothetical control input for the child node; where k represents the current time.

[0025] D3: At the initial sampling time, assuming that all car sub-nodes of the train are running at the same speed, initialize the control input and output trajectory of the car node;

[0026] D4: Based on the current state, the car's own output trajectory, and the assumed output trajectory of neighboring car sub-nodes, set the control input as follows: The control sequence for step length, and the state information are: A sequence of state variables of step length;

[0027] D5: Predict the changes in the state variable sequence under the influence of the control sequence based on the prediction model, evaluate the quality of the control sequence behavior based on the optimization objective, solve the optimization problem using linear programming, and calculate the optimal control sequence at the current time point.

[0028] D6: The control sequence obtained from the optimization solution is used as the hypothetical control input, and the future state variable sequence predicted based on the hypothetical control input is used as the hypothetical output trajectory. The hypothetical control input and hypothetical output trajectory are passed to the neighboring nodes, and information from the neighboring child nodes is received.

[0029] D7: Calculate a reference value for the consistency speed between child nodes based on the information passed by neighboring child nodes and the information of the current child node;

[0030] D8: Extract the first item of the control sequence as the optimal control quantity at the current time point and apply it to the system. Based on the control quantity at the current time point, update the system state estimate and advance the time to the next time point.

[0031] D9: Repeat D4~D8 until the vehicle comes to a complete stop.

[0032] An electronic device, comprising:

[0033] One or more processors;

[0034] Storage device for storing one or more programs.

[0035] When the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned method.

[0036] A computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method.

[0037] The development of this invention was partially supported by the National Natural Science Foundation of China (62172448) and the major industry-university collaborative project "Research and Application of High-Density Automatic Operation Control Technology for Self-Organizing Networks of Heavy-Haul Freight Trains". The technical advantages of this invention are that, by incorporating the force-motion relationship of multiple train carriages and using distributed Franklin model predictive control, it avoids the problem of inaccurate modeling and control input caused by neglecting some nonlinear elements during modeling. Simultaneously, this invention considers the nonlinear coupling between adjacent carriages. By constructing a consistent braking target for the distributed braking system of the train, it ensures that sub-nodes can obtain some state information from other nodes, verifying whether the consistent braking target has been achieved, and serving as the basis for designing the optimal control strategy. Through distributed model predictive control, the optimal distributed braking strategy of the train is calculated, which can adjust the braking force of each carriage individually during train braking, reducing the magnitude of the interaction force between carriages, thereby reducing the speed inconsistency of carriages during braking, extending the lifespan of train carriages and connecting devices, and ensuring the safety and ride comfort of the vehicle during braking. Attached Figure Description

[0038] Figure 1 This is a flowchart of the method for coordinated communication consistency braking control of heavy-load combined trains under braking scenarios according to the present invention;

[0039] Figure 2 This is a block diagram of the overall structure of the braking control system provided in an embodiment of the present invention;

[0040] Figure 3 This is a structural diagram of the node communication topology model provided in the embodiments of the present invention;

[0041] Figure 4 This is a block diagram of the network-physical model structure of a train multi-carriage provided in an embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the braking system control according to an embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram of the structure of the train distributed braking system in the embodiment of the invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] In the embodiment of the method and device for coordinated communication and consistent braking control of heavy-haul combined trains under the braking scenario of the present invention, the flowchart of the method and device for coordinated communication and consistent braking control of heavy-haul combined trains is as follows: Figure 1 As shown in the figure, the method and device for coordinated communication-based braking control of heavy-load combined trains under braking scenarios includes the following steps:

[0046] Step S01: Obtain information about each train car, including speed, acceleration mass, braking force, friction force, and coupling force between cars. In this embodiment, the connection between adjacent cars is considered as an elastic mechanical connection. The coupling force generated by the relative displacement between cars is called the coupler force; friction force is the rolling mechanical resistance between the car wheels and the track. In this step, each car is considered as a separate point mass, fully considering the coupling force between each car and its adjacent cars, making it easier to perform distributed control of the car braking force.

[0047] Step S02: Based on the state information of each carriage, describe the influence of various forces on the carriage speed and acceleration. In this step, aerodynamic drag and additional drag caused by changing road conditions are further considered, and the relationship between various forces and the carriage motion state is described as follows:

[0048]

[0049] in , , These represent the mass, speed, and geographical location of the i-th carriage, where i = {1, 2, ..., n}. express The derivative of , where t represents time. Let the braking force of the i-th carriage be... Let i be the magnitude of the coupling force between the i-th car and the (i+1)-th car. The derivative of v is the acceleration. Let be the frictional force acting on the i-th carriage. The i-th car is subjected to additional resistance caused by external factors, including track incline, curvature, and tunnel. This represents aerodynamic drag, also known as wind resistance. In this embodiment, only the wind resistance of the first carriage is considered.

[0050] Step S03: Design the consistent braking objective of the train's distributed braking system. In this step, the connection between adjacent train cars can be considered as an elastic mechanical connection. The coupling force generated by the relative displacement is called the coupler force. The upper limit of the coupler force is determined by the material structure of the coupler itself. To ensure the safety of train operation, the coupler force is strictly constrained within a bounded safety range. Furthermore, during the process of the speed of each heavy-load combined train car node converging to the desired speed curve, the speed difference and relative displacement of adjacent heavy-load combined train car nodes are kept as close as possible to a small neighborhood of zero.

[0051] Step S04: Based on the topological relationship between the vehicles, define the communication topology between the train carriages; in this step, the communication topology structure is used to characterize whether the carriages can exchange their respective motion information and force conditions.

[0052] Step S05: According to the communication topology in step S04, the carriage obtains the status information of other carriages. In this step, according to the inter-node communication topology, the carriage communicates with other carriages via MVB bus and other means to obtain information such as the position, speed, and acceleration of other carriages, which facilitates the design of a distributed model predictive controller.

[0053] Step S06: Based on the braking target in step S03 and the carriage state information in step S05, a distributed model predictive controller is designed to allocate braking force to each carriage. In this step, the optimal control input sequence at the current time step is calculated by solving an optimization problem. Various optimization algorithms can be used to solve the optimization problem, such as linear programming, quadratic programming, or nonlinear programming.

[0054] The optimal control input for the current time step is extracted from the optimal control input sequence and applied to the system. The system executes this control input to generate the system state for the next time step.

[0055] The system's state estimate is updated by measuring the system's output and using state estimation algorithms. This can be done by filtering, estimating, or updating the state based on observation data and the system model.

[0056] Continue calculating the next time step and repeat the above steps to optimize the control input for the next time step.

[0057] In this process, further including

[0058] S61: Using the Franklin model to predict the future dynamic behavior of each car in a heavily loaded combined train as a child node, the prediction time domain is set as follows: Control time domain is ;

[0059] S62: Based on the prediction time domain set in S61 Control time domain For prediction in the time domain Define the predicted output trajectory, optimal output trajectory, and assumed output trajectory of the child nodes for the control time domain. Define the predictive control input, optimal control input, and hypothetical control input for the child node; where k represents the current time.

[0060] S63: At the initial sampling time, assuming that all car sub-nodes of the train are running at the same speed, initialize the control input and output trajectory of the car node;

[0061] S64: Based on the current state, the car's own output trajectory, and the assumed output trajectory of neighboring car sub-nodes, the control input is set as follows: The control sequence for step length, and the state information are: A sequence of state variables of step length;

[0062] S65: Predict the changes in the state variable sequence under the influence of the control sequence based on the prediction model, evaluate the quality of the control sequence behavior based on the optimization objective, solve the optimization problem using linear programming, and calculate the optimal control sequence at the current time point.

[0063] S66: The control sequence obtained from the optimization solution is used as the hypothetical control input, and the future state variable sequence predicted based on the hypothetical control input is used as the hypothetical output trajectory. The hypothetical control input and hypothetical output trajectory are passed to the neighboring nodes, and information from the neighboring child nodes is received.

[0064] S67: Calculate a reference value for the consistency speed between child nodes based on the information passed by neighboring child nodes and the information of the current child node;

[0065] S68: Extract the first item of the control sequence as the optimal control quantity at the current time point and apply it to the system. Based on the control quantity at the current time point, update the system state estimate and advance the time to the next time point.

[0066] S69: Repeat S64~S68 until the vehicle comes to a complete stop.

[0067] See Figure 5 The method in this embodiment is used to control such as Figure 5The braking system comprises carriage sensors, an MVB bus system, a distributed model predictive controller, and a distributed braking system. The distributed braking system, the controlled object of this patent, is used to achieve consistent braking of the train. It achieves this by applying different pressures to the train pipes of different carriages, thus providing different braking forces to each carriage. The carriage sensors measure the carriage's state information in real time, primarily including speed, acceleration, and position information. The MVB bus system facilitates communication between the different carriages' state information. The distributed model predictive controller's main functions include receiving state information, running an algorithm to estimate real-time model parameters, running a predictive algorithm to predict the future state information of the carriages, solving a multi-objective optimization problem, and outputting the optimal control law to the distributed braking system.

[0068] In this embodiment, a schematic diagram of the distributed braking system is shown below. Figure 6 As shown, it further includes components such as a pre-control air cylinder pressure control module, a balancing air cylinder pressure control module, and a balancing air cylinder. The distributed braking system switches between the balancing air cylinder pressure control component and the pre-control air cylinder pressure control component to control the balancing air cylinder pressure according to normal operating conditions and ER backup operating conditions; the ER / BP relay valve uses the balancing air cylinder as a reference pressure to control the air supply to the train pipe in the main air direction, thereby realizing the indirect control of the train pipe by the balancing air cylinder.

[0069] In this embodiment, the vehicle compartment sensor further includes: a speed sensor, an acceleration sensor, and a position sensor. The speed sensor is used to collect the vehicle compartment's travel speed; the acceleration sensor is used to collect the vehicle compartment's current acceleration value; and the position sensor is used to measure the distance between vehicles.

[0070] In this embodiment, the control system further includes a computer and a DSP control board. The computer is used to run algorithms to estimate real-time model parameters, run prediction algorithms to predict the future state information of the hybrid energy storage system, solve multi-objective optimization problems, and read parameter data from the database. The DSP control board is used to receive the optimal setpoints solved by the computer, implement negative feedback control, and output control quantities to the bidirectional distributed braking system.

[0071] In summary, in this embodiment, the speed, acceleration, and position information collected by the carriage sensors are transmitted to the computer via the MVB bus, and the computer runs the data sequentially. The DSP control board executes the optimal braking system control, achieving fast and efficient train-wide consistent braking control, improving safety during train braking, and extending the service life of the carriage connection mechanism.

[0072] According to embodiments of the present invention, the present invention also provides an electronic device and a computer-readable medium.

[0073] Electronic devices include:

[0074] One or more processors;

[0075] Storage device for storing one or more programs.

[0076] When the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned method.

[0077] Similarly, the computer-readable medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the braking control method of the embodiments of the present invention.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A heavy-haul consist cooperative communication consistent braking control method, characterized in that, Includes the following steps A: Obtain the status information of each carriage of the heavy-haul combined train; B: Treat each carriage as an individual point mass, and based on the state information and other forces acting on the train during operation, describe the influence of each force on the carriage's velocity and acceleration using the Franklin operator. C: Determine the consistency braking target of the train's distributed braking system based on the speed difference and relative displacement between each carriage; D: Based on the influence of each force on the car in step B and the braking target in step C, design a distributed Franklin model predictive control controller to distribute the braking force of each car. Step B includes: Based on the collected state information, and further considering the additional resistance caused by aerodynamic drag and changing road conditions, the relationship between various forces and the motion state of the carriage is expressed as follows: ； in , , These represent the mass, speed, and geographical location of the i-th carriage, where i = {1, 2, ..., n}. express The derivative of , where t represents time. Let the braking force of the i-th carriage be... Let i be the magnitude of the coupling force between the i-th car and the (i+1)-th car. The derivative of v is the acceleration. Let be the frictional force acting on the i-th carriage. The i-th car is subjected to additional resistance caused by external factors, including track incline, curvature, and tunnel. Indicates aerodynamic drag; The relationship between the force and the motion state of the carriage is discretized as follows: ； in, The chi-square represents the state variable, and k represents the current time. Indicates the sampling time interval. , , and These are all intermediate variables used in the calculation, and the calculation method is as follows: ； where f i is a state transition function; In step C, the goal of consistent braking is to keep the speed difference and relative displacement of adjacent carriages within a small neighborhood where they are as close to zero as possible during the process of the travel speed at each heavy-load combined train carriage node converging to the desired speed curve.

2. The method according to claim 1, characterized in that, In step A, the state information of the carriage includes speed, acceleration, mass, braking force, friction force, and coupling force between carriages; wherein the connection method of adjacent carriages of the train is regarded as an elastic mechanical connection, then the coupling force generated by the relative displacement between carriages is called the coupler force; friction force is the rolling mechanical resistance between the carriage wheels and the track.

3. The method according to claim 1, characterized in that, Step D includes: D1: Using the Franklin model to predict the future dynamic behavior of each car in a heavily loaded combined train (as a child node), the prediction time domain is set as follows: Control time domain is ; D2: Based on the prediction time domain set in D1 Control time domain For prediction in the time domain Define the predicted output trajectory, optimal output trajectory, and assumed output trajectory of the child nodes for the control time domain. Define the predictive control input, optimal control input, and hypothetical control input for the child node; where k represents the current time. D3: At the initial sampling time, assuming that all car sub-nodes of the train are running at the same speed, initialize the control input and output trajectory of the car node; D4: Based on the current state, the car's own output trajectory, and the assumed output trajectory of neighboring car sub-nodes, set the control input as follows: The control sequence for step length, and the state information are: A sequence of state variables of step length; D5: Predict the changes in the state variable sequence under the influence of the control sequence based on the prediction model, evaluate the quality of the control sequence behavior based on the optimization objective, solve the optimization problem using linear programming, and calculate the optimal control sequence at the current time point. D6: The control sequence obtained from the optimization solution is used as the hypothetical control input, and the future state variable sequence predicted based on the hypothetical control input is used as the hypothetical output trajectory. The hypothetical control input and hypothetical output trajectory are passed to the neighboring nodes, and information from the neighboring child nodes is received. D7: Calculate a reference value for the consistency speed between child nodes based on the information passed by neighboring child nodes and the information of the current child node; D8: Extract the first item of the control sequence as the optimal control quantity at the current time point and apply it to the system. Based on the control quantity at the current time point, update the system state estimate and advance the time to the next time point. D9: Repeat D4~D8 until the vehicle comes to a complete stop.

4. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-3.

5. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-3.

Images