Virtual marshalling train distributed control method and system based on running trajectory optimization

Abstract

The invention discloses a virtual marshalling train distributed control method and system based on running track optimization, and the method comprises the steps: firstly, considering the resultant force in a train running process, and building a train simple substance point kinetic equation; secondly, in the virtual marshalling state, train operation conditions are divided according to the spatial position in combination with a train driving control strategy; thirdly, by means of the virtual marshalling technology, all train units in marshalling in the train running process are made to be consistent in speed, constraint conditions of safety protection distance and space-time safety protection are set, the train running energy consumption and comfort model is introduced into a multi-target hiking optimization algorithm, and the energy consumption and comfort of the train are optimized in combination with the constraint conditions; optimizing the running track according to the working condition state of the train; and finally, establishing a distributed self-adaptive sliding mode control law of the train according to the optimized running track. By optimizing the running track of the train, an accurate protection strategy is realized, and the control method established based on the optimized track can ensure stable running of the train while improving the railway transportation efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit control, and in particular relates to a distributed control method and system for a virtual marshaling train based on running trajectory optimization. Background Art

[0002] High-speed trains have become an important part of modern transportation with their fast speed, convenience, efficiency, comfort and environmental protection. As an emerging urban rail transit operation organization mode, virtual marshaling technology is flexible, efficient and intelligent, and can significantly improve the operation efficiency and service level of urban rail transit.

[0003] Virtual marshaling can not only significantly shorten the safety interval between trains, thereby improving the line's throughput capacity, but also dynamically adjust the combination of train units without stopping. Under the virtual marshaling operation mode, the operation trajectory optimization aims to generate the optimal operation trajectory based on train tasks, line conditions and external restrictions (such as train speed, energy consumption, safety interval between unit trains, etc.) to ensure safety, improve transportation efficiency, save energy and improve passenger comfort. Compared with traditional centralized control, which manages all train operations through a central system, distributed control decomposes control tasks into multiple train units, and completes operation control with the help of communication and collaboration mechanisms, thereby improving system performance and ensuring safe and smooth operation of trains. Therefore, using advanced control technology and optimization algorithms to ensure the safe and efficient operation of high-speed trains under complex line conditions has become an important research direction and technical focus in the railway field.

[0004] At present, the research on multi-train trajectory optimization widely uses algorithms such as dynamic programming and multi-objective particle swarm optimization. However, these algorithms have a long calculation time and cannot meet the real-time optimization requirements in high-speed train operation scenarios. At the same time, current research is mostly based on ideal environments and rarely involves complex factors such as communication delays or communication failures in actual operation. Summary of the invention

[0005] In view of the problems existing in the above-mentioned background technology, the object of the present invention is to provide a distributed control method and system for virtual train formation based on running trajectory optimization.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: The distributed control method of virtual marshaling train based on running trajectory optimization includes the following steps: S1. Considering the resultant force during the train operation, including traction, braking force, time-varying basic resistance and additional resistance, the train single-particle dynamic equation is established; S2: In the virtual marshaling state, the whole process of train operation is divided into traction condition, cruising condition, coasting condition and braking condition according to the spatial position and the train driving operation strategy; S3. Through virtual marshaling technology, all train units in the marshaling process can achieve the same speed during train operation, set the same safety protection distance between trains, and set time and space safety protection constraints, so as to ensure the safety of train operation in virtual marshaling while ensuring the efficient operation of trains; S4. Establishing the comfort model of train travel: , In the formula, For trains exist Comfort index at all times, For trains exist The acceleration of time, is the number of discrete sampling points in the time period, RMSE represents the root mean square error; S5. When the train is in different working conditions, the energy consumption and comfort model of the train operation is introduced into the Multi-Objective Hiking Optimization Algorithm (MOHOA). Combined with the constraints of spatiotemporal safety protection, according to the working state of the train, the acceleration, speed, braking force and error feedback of the train are considered to optimize the running trajectory of the train. S6. Establishing the distributed adaptive sliding mode control law of the train; Firstly, according to the train single-particle dynamics equation and the optimized train running trajectory, the train single-particle error dynamics model is constructed: , In the formula, , Train exist The displacement error and velocity error at the moment, and Respectively represent trains exist The actual displacement and velocity at the moment, and Respectively represent trains exist The desired displacement and speed at each moment; Secondly, the tracking error sliding surface is designed according to the train single-particle error dynamics model : , In the formula, is the sliding surface parameter; Thirdly, considering the train parameters and external disturbances, the adaptive sliding mode control law is constructed by combining the tracking error sliding surface : , in, is the equivalent control law, For the switching control law, we have: , In the formula, For interference, For the Train in The running acceleration at the moment, , All are switching gains; Finally, we build the saturation function : , In the formula, As a constant, take =0.05; use replace , then the adaptive sliding mode control law The expression is: .

[0007] Furthermore, in step S1, the train single-particle dynamics equation is as follows: , In the formula, is the mass of the train; for The location of the train at the time, for The speed of the train at the time, for Train acceleration at the moment; , , are the traction force, braking force and time-varying resistance of the train, respectively. The time-varying resistance includes the time-varying basic resistance and additional resistance. It is a coefficient that adjusts the running state of the train and affects the relationship between the traction force on the train and the mass of the train. It is the coefficient that adjusts the external resistance of the train, and affects the relationship between the braking force on the train and the mass of the train; and The values ​​are as follows: .

[0008] Furthermore, the resultant force during the operation of the train satisfies the following formula: , In the formula, and are the time-varying basic resistance and additional resistance, , , They are the curve additional resistance, slope additional resistance and tunnel additional resistance of the line where the train is located; The nominal known part of the basic resistance is expressed by Davis equation: , In the formula, , , It is the empirical parameter of Davis equation and can be calculated by wind tunnel test; The curve additional resistance, ramp additional resistance and tunnel additional resistance are expressed as: , In the formula, is the curve radius of the line where the train is located, is the slope of the line on which the train is located, is the acceleration due to gravity; is the length of the tunnel on the line where the train is located.

[0009] Furthermore, in step S3, the safety protection distance is: , In the formula, It is the safety protection distance of the train; The distance required for the train to brake; Train travel distance within system response time; For additional safety distance.

[0010] Furthermore, in step S3, the spatiotemporal security protection includes time security protection and space security protection, and the time security protection satisfies the following constraints: , In the formula, for The position of the vehicle in front at the moment; For the train conductor; for The position of the vehicle behind at that moment; is the actual running time of the train; The specified train running time is It indicates that the actual train running time is approaching the prescribed train running time; Space security protection meets the following constraints: , In the formula, is the running distance of the train within the reaction time of the preceding train system; is the train running distance within the reaction time of the following vehicle system; The distance required for the vehicle ahead to brake; The distance required for the rear vehicle to brake.

[0011] Furthermore, in step S5, the energy consumption of the train operation is calculated as follows: , In the formula, is the train running energy consumption, Indicates auxiliary power, It is the conversion factor of electrical energy into mechanical energy under the traction state of the train; It is the conversion factor of mechanical energy into electrical energy when the train is in braking state.

[0012] Furthermore, in step S5, the method of optimizing the train running trajectory using the multi-objective walking optimization algorithm is as follows: S51, a multi-objective walking optimization algorithm reads the line data and speed limit of the train operation; S52. According to the actual operation requirements, the multi-objective walking optimization algorithm sets a series of train operation constraints and selects the final conditions for iteration; S53, the multi-objective walking optimization algorithm simulates the running trajectory of the train, and generates a Pareto frontier solution set according to the energy consumption, comfort index and time of the train in different states, and stores it in the external reserve set; S54, the multi-objective walking optimization algorithm uses the global search strategy of the population to guide the subsequent trajectory optimization. By selecting the optimal solution in the current state, it guides the population to develop in the optimization direction and gradually approaches the optimal solution set; S55, by updating the speed and position of the train in real time, adjusting the running state of the train according to the local optimal solution, performing train traction calculation, and maintaining the tracking distance, thereby ensuring that the train continues to track the optimal trajectory during operation; S56. Determine whether the running state of the train satisfies the constraints and the final conditions of the iteration. If not, repeat steps S53-S55. If yes, store the optimal curve, output the solution set at this time as the final train running trajectory, and output the corresponding energy consumption and comfort indicators.

[0013] The present invention further provides a virtual marshaling train distributed control system based on running trajectory optimization, comprising: Information acquisition module: obtains the line information and movement authorization information of the train operation, as well as the operation status information shared by the train; Communication module: transmits the line information of the train operation to the information processing module of the first train in real time and enables it to receive the mobile authorization information. The operation status information shared by the preceding train is transmitted to the information processing module of the subsequent train in real time. The information processing module of the subsequent train adjusts the operation strategy according to the preceding train and the provided mobile authorization information; Running track optimization module: receives the train running track transmitted by the information processing module, and optimizes the running track of the first train under the virtual marshaling through the built-in multi-objective walking optimization algorithm; Tracking control module: Adopts adaptive sliding mode control method to track the subsequent trains on the optimized trajectory of the first train, and transmits the tracking operation results to the on-board ATP.

[0014] Compared with the shortcomings and deficiencies of the prior art, the present invention has the following beneficial effects: The present invention uses a multi-objective walking optimization algorithm to obtain train position, speed, running time and status information in real time, thus realizing an accurate protection strategy. Trajectory tracking methods such as adaptive sliding mode control can improve railway transportation efficiency while ensuring smooth train operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of a distributed control method for a virtual marshaling train based on running trajectory optimization provided by an embodiment of the present invention; Figure 2 It is a flowchart of optimizing the running track of a high-speed train using MOHOA provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the design principle of the distributed adaptive sliding mode control for high-speed trains provided by an embodiment of the present invention; Figure 4 It is a structural block diagram of a virtual marshaling train distributed control system based on running trajectory optimization provided by an embodiment of the present invention; Figure 5 is a comparison diagram before and after the optimization of the high-speed train running track provided by an embodiment of the present invention; Figure 6 is a schematic diagram of high-speed train displacement tracking provided by an embodiment of the present invention; Figure 7 It is a schematic diagram of high-speed train speed tracking provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0017] Combination Figure 1 The flowchart of the distributed control method of virtual train formation based on running trajectory optimization is shown to illustrate the present invention.

[0018] 1. According to the analysis of the dynamic principles of high-speed trains and the resultant forces during train operation, it includes traction, braking force, time-varying basic resistance and additional resistance. The time-varying resistance includes the time-varying basic resistance and additional resistance. The additional resistance includes the curve additional resistance, slope additional resistance and tunnel additional resistance of the line where the train is located.

[0019] The resultant force during train operation satisfies the following formula: (1), In the formula, is the train mass, for The train acceleration at that moment, , , are the traction force, braking force and time-varying resistance of the train respectively. and They are the time-varying basic resistance and additional resistance. The additional resistance is related to the line environment where the train runs. , , They are the curve additional resistance, slope additional resistance and tunnel additional resistance of the line where the train is located.

[0020] Based on Newton's second law, the train single particle dynamics equation is established: (2), In the formula, for The location of the train at the time, for The speed of the train at the time. is the coefficient for adjusting the train running status, is the coefficient for adjusting the external resistance of the train. and The values ​​are as follows: (3).

[0021] The basic resistance of high-speed trains is mainly composed of mechanical resistance, friction resistance, aerodynamic resistance, etc., which is related to the train structural parameters, running speed and environment. The nominal known part of the basic resistance is expressed by the Davis equation as follows: (4), In the formula, , , It is the empirical parameter of Davis equation and can be calculated by wind tunnel test.

[0022] The additional resistance of curves, ramps and tunnels is expressed as: (5), In the formula, is the curve radius of the line where the train is located, is the slope of the line where the train is located, and the additional resistance per unit mass of the ramp is numerically equal to the thousandth value of the slope; is the acceleration due to gravity; is the length of the tunnel on the line where the train is located.

[0023] 2. Combined with the train driving operation strategy, the whole process of train operation is divided into traction condition, cruising condition, coasting condition and braking condition.

[0024] The train operation driving strategy is divided into multiple stages, each stage includes the state variable space and the process of the train running in the section after departure; the whole process of the train running in the section includes traction condition, cruising condition, coasting condition and braking condition; among them, under traction condition, the traction of the train plays a major role, the train is in the acceleration stage, so that the train reaches the required speed; under cruising condition, the stage where the train runs at a constant speed, when the train reaches the required cruising speed, the power system output is balanced with the driving resistance, and the speed remains unchanged; under the inertial condition, the train relies on inertial sliding in an unpowered state, at this time the train power system does not provide traction, the train is only affected by air resistance, rolling resistance and slope resistance, and gradually decelerates; under braking condition, the stage where the train decelerates or stops, braking can be achieved through electric braking (regenerative braking) or mechanical braking. The trajectory optimization algorithm optimizes the trajectory for each stage according to the needs of different working conditions to ensure that the dynamic constraints, energy efficiency requirements and comfort requirements are met at each stage.

[0025] 3. During train operation, virtual marshaling technology is used to achieve speed consistency for all train units in the marshaling, and the same safety protection distance is set between trains. By combining time-space safety protection with time-space trajectory optimization, not only the operation safety of trains in virtual marshaling is guaranteed, and sufficient safety distance is always maintained between trains, but also the switching points under different working conditions are accurately optimized through the trajectory optimization algorithm to ensure smooth transition of each working condition. The trajectory optimization algorithm fine-tunes the transition between various working conditions according to the dynamic characteristics of the train, the dynamic response during the working condition conversion, and the safety requirements, minimizes energy consumption, and improves passenger comfort. Specifically, under different working conditions such as traction, cruising, idling and braking, the trajectory optimization algorithm will optimize the motion trajectory of the train by considering the acceleration, speed, braking force and error feedback of the train to ensure smooth and efficient operation. Time-space trajectory optimization not only ensures that the virtual marshaling trains operate in an efficient manner while ensuring safety, but also improves the operating efficiency of the entire system. Finally, through this comprehensive strategy, the train can achieve efficient and smooth operation while ensuring safety, maximizing energy saving and comfort.

[0026] The tracking operation between train formations adopts the relative braking distance mode, and the minimum safe tracking distance between adjacent trains is: (6), In the formula, It is the safety protection distance of the train; The distance required for the train to brake; Train travel distance within system response time; For additional safety distance.

[0027] The spatiotemporal safety protection between trains includes time safety protection and space safety protection. For space, the most unfavorable condition of the previous train running curve is used as the protection basis, and the train is protected from the spatial dimension. By solving the safety distance required for train collision avoidance, the train overspeed protection curve is quickly calculated to ensure the safe operation of the train, that is, the space safety protection meets the following constraints: (7), In the formula, is the running distance of the train within the reaction time of the preceding train system; is the train running distance within the reaction time of the following vehicle system; The distance required for the vehicle ahead to brake; The distance required for the rear vehicle to brake.

[0028] In terms of time, when the train is braking in the most unfavorable way, it is necessary to ensure that the rear train will not collide with the front train within the braking time range of the front train, that is, at the same time When the front and rear trains do not collide, the overall running time approaches the specified running time and fluctuates within the allowed time range, that is, the time safety protection meets the following constraints: (8), In the formula, for The position of the vehicle in front at the moment; For the train conductor; for The position of the vehicle behind at that moment; is the actual running time of the train; The specified train running time is It means that the actual train running time is close to the specified train running time.

[0029] 4. The energy consumed by the train during operation mainly includes train traction, carriage lighting and various electromechanical equipment. The energy consumption calculation of the present invention mainly considers the energy consumption generated by the traction work during the train operation. Since different operating conditions will produce different energy consumption, under the premise that the section length and driving time remain unchanged, the energy consumption value of the train is directly related to the operation strategy. The energy consumption calculation formula for train operation is as follows: (9), In the formula, is the train running energy consumption, Indicates auxiliary power, It is the conversion factor of electrical energy into mechanical energy under the traction state of the train; It is the conversion factor of mechanical energy into electrical energy when the train is in braking state.

[0030] The train should run smoothly during operation to ensure the comfort of passengers. The train control layer will control the train to track the target speed curve, but there will be a certain error between the actual speed and the target speed. In order to ensure accurate speed tracking, the train will frequently switch controllers under certain control strategies, resulting in reduced passenger comfort. What affects the ride comfort is the rate of change of the train's longitudinal acceleration. Therefore, a comfort model for train travel is established: (10), In the formula, For trains exist Comfort index at all times, For trains exist The acceleration of time, is the number of discrete sampling points in the time period. RMSE is a common indicator used to measure the difference between a model's prediction results and the actual observed values. The smaller the root mean square error, the smaller the error between the prediction results and the actual data, indicating that the prediction effect of the model is better.

[0031] 5. When the train is in different working conditions, the energy consumption and comfort model of the train operation is introduced into the Multi-Objective Hiking Optimization Algorithm (MOHOA). Combined with the constraints of time and space safety protection, according to the working conditions of the train, the train's acceleration, speed, braking force and error feedback are considered to optimize the train's running trajectory. Set a reasonable running time range and introduce a penalty mechanism to ensure that the train arrives on time within the specified time; according to the working conditions of the train, the train running trajectory is optimized through MOHOA. The flowchart is referenced Figure 2 , the specific optimization method is as follows: First, MOHOA will read the line data of the train operation, including the specific information of the route (such as track type, starting and ending points, etc.) and the speed limit. These data provide the basic framework for the subsequent trajectory optimization. Then, according to the actual operation requirements, MOHOA will set a series of constraints for train operation, such as the maximum acceleration of the train, speed limit, energy consumption limit, etc. These conditions ensure that the optimization results meet the actual limitations of train operation, while also ensuring the effectiveness and safety of trajectory optimization, and select the final conditions for iteration. Secondly, the train operation will simulate the running trajectory of the train and calculate the energy consumption, comfort index and time under different states. By calculating these indicators, MOHOA can generate a Pareto frontier solution set, which represents the balance between goals such as energy consumption and comfort, and is stored in the external reserve set. After that, MOHOA uses the global search strategy of the population to guide the subsequent trajectory optimization. By selecting the optimal solution in the current state, it guides the population to develop in the optimization direction and gradually approaches the optimal solution set. Then, by updating the speed and position of the train in real time, adjusting the running state of the train according to the local optimal solution, performing train traction calculations, and maintaining the tracking distance, the goal of this step is to make the train run smoother and meet the predetermined comfort and energy consumption standards. Finally, determine whether the running state of the train meets the constraints and the final conditions of the iteration. If not, regenerate the Pareto frontier solution set and repeat the above steps; if so, the optimization results will be saved, the optimal curve will be stored, and the solution set at this time will be output as the final train running trajectory, and the corresponding energy consumption and comfort indicators will be output at the same time. These optimized trajectories can better balance factors such as train energy consumption, comfort, and running time.

[0032] 6. According to the optimized running trajectory, design the distributed adaptive sliding mode control law of the virtual marshaling high-speed train. The design principle of adaptive sliding mode control is as follows: Figure 3 As shown in the figure, the sliding surface and distributed control law are designed according to the dynamic model of the train to ensure that the train can achieve efficient and stable operation under different working conditions. Traditional sliding mode control may have chattering phenomenon, which affects the actual control effect. Therefore, an adaptive sliding mode control method is introduced to effectively reduce chattering by adjusting the sliding mode parameters in real time.

[0033] According to the train single-particle dynamics equation and the optimized train running trajectory, the train single-particle error dynamics model is constructed: (11), In the formula, , Train exist The displacement error and velocity error at the moment, and Respectively represent trains exist The actual displacement and velocity at the moment, and Respectively represent trains exist The desired displacement and velocity at each moment.

[0034] Design of tracking error sliding surface based on train single-mass error dynamics model : (12), In the formula, is the sliding surface parameter.

[0035] Construct an adaptive sliding mode control law considering train parameters and external disturbances : (13), In the formula, is the equivalent control law, is the switching control law, For interference, For the Train in The running acceleration at the moment, , are all switching gains; then: (14), The function is: (15), Among them, due to Discontinuity causes obvious chattering of the control law, so a saturation function is proposed : (16), In the formula, As a constant, take =0.05; use replace , then the adaptive sliding mode control law The expression is: (17).

[0036] 7. Use Lyapunov stability theory to analyze the effectiveness of the virtual marshaling train operation trajectory optimization and distributed control method of the present invention, and verify whether the system state tends to the equilibrium point under the action of adaptive sliding mode control from the two aspects of equivalent control and switching control, to ensure that it converges to the desired trajectory on the sliding surface in the most stable way. Variables, in order to facilitate the derivation of formulas, time is omitted below .

[0037] about The Lyapunov function is: (18), in, , , Represents different system state variable parameters, such as speed, position, acceleration, etc. , , They are , , The error represents the gap between the actual value and the expected value of the system state variable; Deriving equation (18) and substituting the adaptive sliding mode control law of equation (17) into it, we can obtain: (19), In the formula, for Derivatives of functions; Indicates the sliding surface Derivation; , , is the actual value of the system state variable, usually representing the actual control variable of the system; To control the gain, for The derivative of The running acceleration of the train; for The derivative of The running acceleration of the train; , , They are , , The derivative of , where: (20), In the formula, , , is the expected value of the system state variable. Combining equations (18), (19) and (20), we get: (twenty one), in, , , They are , , The control law in the formula is achieved by switching the gain , Adjust the intensity of error feedback to control variables such as the train's trajectory, speed, and acceleration to ensure smooth and efficient operation of the system. , D is the lumped disturbance, so ,but: (twenty two), in, , , Respectively with , , The deviation associated with the variable represented by . Let: (twenty three), In the formula, represents the overall deviation of the system, then: (twenty four), (25), In the formula, is a parameter related to the acceleration or deceleration of the train; Formulas (24) and (25) can be simplified to: (26), When equation (26) is equal, we can get: (27), In the formula, It's time The time step in the process can be converted into equation (26) by combining equation (27): (28), when When , combined with formula (28), we can get: (29), (30), (31); In summary, by choosing appropriate design parameters, the system error will converge to a region near zero, so the Lyapunov function It will not diverge to infinity, which proves the stability of this method.

[0038] 8. The structural block diagram of the distributed control system of virtual train formation based on running trajectory optimization is as follows Figure 4 As shown, including: Information acquisition module: obtains the line information and movement authorization information of the train operation, as well as the operation status information shared by the train; Communication module: transmits the line information of the train operation to the information processing module of the first train in real time and enables it to receive the mobile authorization information. The operation status information shared by the preceding train is transmitted to the information processing module of the subsequent train in real time. The information processing module of the subsequent train adjusts the operation strategy according to the preceding train and the provided mobile authorization information; Running track optimization module: receives the train running track transmitted by the information processing module, and optimizes the running track of the first train under the virtual marshaling through the built-in multi-objective walking optimization algorithm; Tracking control module: Adopts adaptive sliding mode control method to track the subsequent trains on the optimized trajectory of the first train, and transmits the tracking operation results to the on-board ATP.

[0039] 9. Verification Example By obtaining the information of four CRH5 trains of a high-speed railway, the original speed, displacement and time data of the four trains are obtained, and the average value of the corresponding running data of the four trains is taken. The original unoptimized energy consumption and comfort indexes are calculated, and the running trajectory obtained after optimization is used to calculate the optimized energy consumption and comfort indexes. The average values ​​of the energy consumption and comfort indexes of the four trains before and after the running trajectory optimization are taken respectively. The results are shown in Table 1: Table 1 Comparison of train energy consumption and comfort index before and after running trajectory optimization

[0040] Figure 5Table 1 shows the comparison of high-speed train running trajectory before and after optimization. Figure 5 It can be seen that the optimized train operation energy consumption is better than the original unoptimized train operation energy consumption, and the optimized train operation comfort index is also better than the original unoptimized comfort index.

[0041] Taking the four CRH5 trains of the data source shown in Table 1 as an example, by taking the average of the running energy consumption and comfort indexes of the four trains, the running curves of the positions of the four trains tracking and running over time are obtained by using the virtual marshaling train distributed control method based on running trajectory optimization proposed in the present invention, as shown in FIG. Figure 6 As shown, the running curves of the speed of the four trains changing with time are obtained at the same time, such as Figure 7 As shown. Figure 6 and Figure 7 It can be seen that the distributed adaptive sliding mode control law of the virtual marshaling high-speed train designed in the present invention can realize the tracking operation control of the train and ensure the smooth operation of the train.

[0042] 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 and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A distributed control method for virtual train formation based on running trajectory optimization, characterized in that: The steps include: S1. Considering the resultant force during the train operation, including traction, braking force, time-varying basic resistance and additional resistance, the train single-particle dynamic equation is established; S2: In the virtual marshaling state, the whole process of train operation is divided into traction condition, cruising condition, coasting condition and braking condition according to the spatial position and the train driving operation strategy; S3. Through virtual marshaling technology, all train units in the marshaling process can achieve the same speed during train operation, set the same safety protection distance between trains, and set time and space safety protection constraints; S4. Establishing the comfort model of train travel: ， In the formula, For trains exist Comfort index at all times, For trains exist The acceleration of time, is the number of discrete sampling points in the time period, RMSE represents the root mean square error; S5. When the train is in different working conditions, the energy consumption and comfort model of the train operation is introduced into the multi-objective walking optimization algorithm. Combined with the constraints of spatiotemporal safety protection, according to the working state of the train, the acceleration, speed, braking force and error feedback of the train are considered to optimize the running trajectory of the train. S6. Establishing the distributed adaptive sliding mode control law of the train; Firstly, according to the train single-particle dynamics equation and the optimized train running trajectory, the train single-particle error dynamics model is constructed: ， In the formula, , Train exist The displacement error and velocity error at the moment, and Respectively represent trains exist The actual displacement and velocity at the moment, and Respectively represent trains exist The desired displacement and speed at each moment; Secondly, the tracking error sliding surface is designed according to the train single-particle error dynamics model : ， In the formula, is the sliding surface parameter; Thirdly, considering the train parameters and external disturbances, the adaptive sliding mode control law is constructed by combining the tracking error sliding surface : ， in, is the equivalent control law, For the switching control law, we have: ， In the formula, For interference, For the Train in The running acceleration at the moment, , All are switching gains; Finally, we build the saturation function : ， In the formula, As a constant, take =0.05; use replace , then the adaptive sliding mode control law The expression is: 。 2. The distributed control method for virtual train formation based on running trajectory optimization according to claim 1, characterized in that: In step S1, the train single-particle dynamics equation is as follows: ， In the formula, is the mass of the train; for The location of the train at the time, for The speed of the train at the time, for Train acceleration at the moment; , , are the traction force, braking force and time-varying resistance of the train, respectively. The time-varying resistance includes the time-varying basic resistance and additional resistance. is the coefficient for adjusting the train running status, is the coefficient for adjusting the external resistance of the train. and The values ​​are as follows: 。 3. The distributed control method for virtual train formation based on running trajectory optimization according to claim 2, characterized in that: The resultant force during the train operation satisfies the following formula: ， In the formula, and are the time-varying basic resistance and additional resistance, , , They are the curve additional resistance, slope additional resistance and tunnel additional resistance of the line where the train is located; The nominal known part of the basic resistance is expressed by Davis equation: ， In the formula, , , It is the empirical parameter of Davis equation and can be calculated by wind tunnel test; The curve additional resistance, ramp additional resistance and tunnel additional resistance are expressed as: ， In the formula, is the curve radius of the line where the train is located, is the slope of the line on which the train is located, is the acceleration due to gravity; is the length of the tunnel on the line where the train is located.

4. The distributed control method for virtual train formation based on running trajectory optimization according to claim 1, characterized in that: In step S3, the safety protection distance is: ， In the formula, It is the safety protection distance of the train; The distance required for the train to brake; Train travel distance within system response time; For additional safety distance.

5. The distributed control method for virtual train formation based on running trajectory optimization according to claim 1, characterized in that: In step S3, the spatiotemporal security protection includes time security protection and space security protection, and the time security protection satisfies the following constraints: ， In the formula, for The position of the vehicle in front at the moment; For the train conductor; for The position of the vehicle behind at that moment; is the actual running time of the train; The specified train running time is It indicates that the actual train running time is approaching the prescribed train running time; Space security protection meets the following constraints: ， In the formula, is the running distance of the train within the reaction time of the preceding train system; is the train running distance within the reaction time of the following vehicle system; The distance required for the vehicle in front to brake; The distance required for the rear vehicle to brake.

6. The distributed control method for virtual train formation based on running trajectory optimization according to claim 1, characterized in that: In step S5, the energy consumption calculation formula of the train operation is as follows: ， In the formula, is the train running energy consumption, is the traction force on the train, is the braking force on the train, Indicates auxiliary power, It is the conversion factor of electrical energy into mechanical energy under the traction state of the train; It is the conversion factor of mechanical energy into electrical energy when the train is in braking state.

7. The distributed control method for virtual train formation based on running trajectory optimization according to claim 1, characterized in that: In step S5, the method of optimizing the train running trajectory using the multi-objective walking optimization algorithm is as follows: S51, a multi-objective walking optimization algorithm reads the line data and speed limit of the train operation; S52. According to the actual operation requirements, the multi-objective walking optimization algorithm sets a series of train operation constraints and selects the final conditions for iteration; S53, the multi-objective walking optimization algorithm simulates the running trajectory of the train, and generates a Pareto frontier solution set according to the energy consumption, comfort index and time of the train in different states, and stores it in the external reserve set; S54, the multi-objective walking optimization algorithm uses the global search strategy of the population to guide the subsequent trajectory optimization. By selecting the optimal solution in the current state, it guides the population to develop in the optimization direction and gradually approaches the optimal solution set; S55, by updating the speed and position of the train in real time, adjusting the running state of the train according to the local optimal solution, performing train traction calculation, and maintaining the tracking distance, thereby ensuring that the train continues to track the optimal trajectory during operation; S56. Determine whether the running state of the train satisfies the constraints and the final conditions of the iteration. If not, repeat steps S53-S55. If yes, store the optimal curve, output the solution set at this time as the final train running trajectory, and output the corresponding energy consumption and comfort indicators.

8. A system for implementing the distributed control method of virtual train formation based on running trajectory optimization as described in any one of claims 1 to 7, characterized in that: include: Information acquisition module: obtains the line information and movement authorization information of the train operation, as well as the operation status information shared by the train; Communication module: transmits the line information of the train operation to the information processing module of the first train in real time and enables it to receive the mobile authorization information. The operation status information shared by the preceding train is transmitted to the information processing module of the subsequent train in real time. The information processing module of the subsequent train adjusts the operation strategy according to the preceding train and the provided mobile authorization information; Running track optimization module: receives the train running track transmitted by the information processing module, and optimizes the running track of the first train under the virtual marshaling through the built-in multi-objective walking optimization algorithm; Tracking control module: Adopts adaptive sliding mode control method to track the subsequent trains on the optimized trajectory of the first train, and transmits the tracking operation results to the on-board ATP.

Images