Train virtual coupling queue finite time control method and device, medium and product

By combining the global fast terminal sliding mode surface and the new variable-index power approach law, a distributed controller is designed to solve the problem of slow convergence speed and insufficient robustness of train spacing errors in the virtual marshalling queue of high-speed trains, and the train is realized Fast convergence and stability improvement of virtual coupled queues.

CN118915515APending Publication Date: 2024-11-08EAST CHINA JIAOTONG UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410920845.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing high-speed train virtual marshalling queue control method is mainly based on the Lyapunov asymptotic stability of infinite time, which makes it difficult to achieve time optimization of control results, and it is difficult to ensure the rapid convergence of train spacing errors and the robustness of the system within a limited time.

Method used

The distributed controller of the train virtual coupling queue is designed using global fast terminal sliding mode surface and a new variable-index power approach law. By constructing a train distributed dynamic model and a global fast terminal sliding mode surface and power approach law. Distributed controller realizes limited time control of train virtual coupled queues.

Benefits of technology

Improve the convergence performance and strong queue stability of train virtual coupled queues, reduce vibration phenomenon, ensure that train spacing errors quickly converge to the sliding mode surface within a limited time, and improve the robustness and ride comfort of the system. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915515A_ABST
    Figure CN118915515A_ABST
Patent Text Reader

Abstract

The invention discloses a train virtual coupling queue finite time control method and device, a medium and a product, and relates to the field of train virtual coupling queue control, and the method comprises the steps: constructing a train distributed dynamics model and a train queue spacing error model, and determining a train spacing error between each train unit and a corresponding target position; constructing a distributed controller based on a global fast terminal sliding mode surface and a power reaching law; the train running speed and the train distance error of each train unit serve as input, and the finite time control law of each train is output through a distributed controller; and inputting the finite time control law of each train unit into the train distributed dynamical model to realize train virtual coupling queue finite time control. According to the method, the global fast terminal sliding mode surface and the power reaching law are adopted to design the distributed controller of the train virtual coupling queue, and the convergence performance of the train spacing of the train virtual coupling queue and the strong queue stability can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of train virtual coupling queue control, and in particular to a train virtual coupling queue finite time control method, device, medium and product based on global fast terminal sliding mode. Background Art

[0002] With the rapid development of my country's high-speed railway, which will maintain a rapid growth trend for a long period of time in the future, the imbalance of passenger flow in time and space is gradually emerging. Some line sections have greater operating pressure during peak hours, and there is a low utilization rate of line and vehicle resources during flat and low peak hours. Virtual coupling technology achieves efficient train coupling operation through advanced communication technology, and has attracted much attention due to its significant advantages in improving railway transportation capacity.

[0003] The key to realizing virtual coupling of trains is to realize the coordinated operation control and safety protection between train units. In recent years, many scholars at home and abroad have conducted research on the control methods of virtual coupling marshaling queues of high-speed trains. At present, the control algorithms for the application of virtual marshaling queues of high-speed trains mainly include model predictive control multi-agent cooperative control, sliding mode control, adaptive control, etc., as well as fleet following performance optimization control methods based on advanced intelligent algorithms such as reinforcement learning. At present, most of the above studies generally analyze and synthesize fleet controllers based on infinite time Lyapunov asymptotic stability, which makes it difficult to achieve time optimization for control results. Studies have shown that finite-time closed-loop control systems have better robustness and anti-interference performance than asymptotically stable control systems. It is of great significance to design a train virtual coupling queue control system using a finite-time control method. Summary of the invention

[0004] The purpose of the present invention is to provide a finite time control method, device, medium and product for a train virtual coupling queue, which adopts a global fast terminal sliding surface and a power reaching law to design a distributed controller for the train virtual coupling queue, thereby improving the convergence performance of the inter-vehicle distance of the train virtual coupling queue and the strong queue stability.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A train virtual coupling queue finite time control method, the train virtual coupling queue finite time control method comprising:

[0007] Construct a distributed dynamics model of the train;

[0008] Constructing a train queue spacing error model according to the target spacing, and determining the vehicle spacing error between each train unit and the corresponding target position based on the train queue spacing error model;

[0009] Construct a distributed controller based on global fast terminal sliding surface and power reaching law;

[0010] Taking the train running speed of each train unit and the inter-vehicle distance error as input, the distributed controller is used to output a finite time control law of each train;

[0011] The finite-time control law of each train unit is input into the train distributed dynamics model to realize the finite-time control of the train virtual coupling queue.

[0012] A computer device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned method for finite time control of a virtual coupled train queue.

[0013] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for finite-time control of a virtual coupled train queue.

[0014] A computer program product includes a computer program, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned method for finite time control of a virtual coupled queue of trains are implemented.

[0015] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0016] The present invention provides a finite time control method, device, medium and product for a train virtual coupling queue. When constructing a controller, a global fast terminal sliding mode control surface (GFTSMS) is introduced. By adjusting the controller parameters, the train spacing error can be placed on the sliding mode surface at the initial moment, thereby accelerating the convergence speed of the vehicle spacing error and maintaining global robustness. The designed convergence law can make the vehicle spacing error converge to the sliding mode surface within a finite time, reducing the chattering phenomenon. In the present invention, a distributed controller for a train virtual coupling queue is designed by combining a global fast terminal sliding mode surface and a power convergence law, which can improve the convergence performance of the vehicle spacing of the train virtual coupling queue and the strong queue stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1A schematic flow chart of a finite time control method for a train virtual coupling queue provided in Example 1 of the present invention;

[0019] Figure 2 A schematic diagram of a train virtual coupling queue control framework provided in Example 1 of the present invention;

[0020] Figure 3 A schematic diagram of a virtual train marshaling operation provided in Example 1 of the present invention;

[0021] Figure 4 A schematic diagram of the force condition per unit mass of a single-particle model of train i provided in Example 1 of the present invention;

[0022] Figure 5 A schematic diagram of a displacement speed time curve of a pilot train provided in Example 1 of the present invention;

[0023] Figure 6 A schematic diagram of a train position time curve (LF) provided in Example 1 of the present invention;

[0024] Figure 7 A schematic diagram of a curve (LF) of a distance change between a train and a preceding vehicle provided in Example 1 of the present invention;

[0025] Figure 8 A schematic diagram of a train speed time curve (LF) provided in Example 1 of the present invention;

[0026] Fig. 9 A schematic diagram of a train speed position time curve (LF) provided in Example 1 of the present invention;

[0027] Fig.10 A schematic diagram of a train position time curve (PF) provided in Example 1 of the present invention;

[0028] Fig.11 A schematic diagram of a curve (PF) of a distance between a train and a preceding vehicle provided in Example 1 of the present invention;

[0029] Fig.12 A schematic diagram of a train speed time curve (PF) provided in Example 1 of the present invention;

[0030] Fig.13 A schematic diagram of a train speed position time curve (PF) provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] The purpose of the present invention is to provide a finite-time control method, device, medium and product for a virtual coupled queue of trains. Aiming at the control problem of the virtual coupled queue of high-speed trains, a finite-time cooperative control strategy for the virtual coupled queue of trains based on the global fast terminal sliding mode control is designed based on the global fast terminal sliding mode surface (GFTSMS) and the new variable exponential power reaching law (VEPRL). Among them, (1) a global fast terminal sliding mode surface (GFTSMS) containing the train spacing error is constructed to accelerate the convergence speed of the train spacing error of the coupled queue and realize the rapid tracking of the target speed. (2) A new variable exponential power reaching law with continuous smoothness is introduced to eliminate the steady-state convergence error of the train spacing caused by external interference, so as to solve the disadvantage that the train spacing error can only converge to a bounded area due to external interference when reaching the steady state. At the same time, due to the continuity of VEPRL, the frequent acceleration and deceleration of the train caused by the jitter when the sliding mode state reaches the sliding mode surface is avoided.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Example 1

[0035] like Figure 1 and Figure 2 As shown, in order to solve the problem of train virtual coupling queue control, the present invention proposes the following Figure 2 The control framework shown. First, a queue transverse error model is constructed based on the expected spacing between trains; then, a distributed longitudinal controller of the train virtual coupling queue is designed based on GFTSMC-VEPRL to achieve rapid convergence of the error. The power exponential reaching law (or the improved power exponential reaching law to eliminate the steady-state convergence error of the vehicle spacing caused by external interference) is introduced. Finally, according to the control input generated by the distributed controller, the train queue is enabled to achieve virtual coupling and expected vehicle spacing operation. A train virtual coupling queue finite time control method in this embodiment, the method includes:

[0036] S1: Construct a distributed dynamics model of the train.

[0037] The present invention uses a single-particle model to study the tracking control problem of virtual coupled train queues on straight tracks. The "leader-follower" scheme is applied. The head car in the virtual coupled queue acts as the leader, runs according to the speed displacement curve set by the control center, and transmits its own running status information to the following car in real time through wireless communication. The following car in the queue is regarded as a follower, and the control strategy of the car is determined by the exchanged information, thereby controlling its own operation.

[0038] For the virtual marshaling topology, the virtual coupled train platoon is a special longitudinal formation system. For different control objectives, the platoon system has different topological structures. Different structures also determine the input of the controller, and then determine the control structure and whether the control effect in the control objective can be achieved. Consider a platoon consisting of a leader and several followers, such as Figure 3 As shown, where p i ,v i Respectively represent the position and speed information of the i-th (i=0,1,2,…,n) train. If the trains in the platoon can obtain the position and speed information of the leader train through vehicle-to-vehicle communication, this information can be regarded as the target speed and position of the entire virtual coupled platoon. At this time, the platoon communication mode is the leader following (LF) structure, such as Figure 3 As shown in (a) in the figure, the overall control goal can be achieved by obtaining the target information of the pilot train; if the trains in the convoy only obtain the position and speed information of the train in front through vehicle-to-vehicle communication, the convoy communication mode is the predecessor following (PF) structure, as shown in Figure 3 As shown in (b), the local control objective is achieved by minimizing the difference between adjacent unit trains.

[0039] In order to establish the optimal control model of the virtual coupled train platoon, the present invention makes the following assumptions:

[0040] (1) The platform length meets the needs of multiple trains stopping at the station, and the train automatic control system meets the high real-time requirements of information transmission, that is, there is no delay or signal packet loss in vehicle communication.

[0041] (2) Multiple trains in a virtual coupled queue run on a single line, without considering the formation and disorganization of trains in the virtual coupled queue at the intersection of the lines during operation.

[0042] (3) Ignore the relevant constraint disturbances caused by factors such as weather and passenger flow during train operation.

[0043] For the control target, when the train virtual coupling queue is running, the actual spacing between the train units needs to be controlled to be greater than the minimum spacing, and at the same time close to the expected spacing, so that the train virtual coupling queue maintains a stable formation. Therefore, the control target of queue stability is divided into local control targets and overall control targets. The local control target is to reduce the speed difference between adjacent unit trains and maintain a safe spacing, and the overall control target is to ensure that all unit trains run stably at the expected spacing. Therefore, the control target of the present invention is:

[0044] Objective 1, local stability: Consider the finite time convergence of the relative spacing between trains, that is, ensure that within a finite time, the spacing error of the system converges to zero, and at the same time, the speed of each follower train can track the speed of the leader train, expressed as:

[0045]

[0046] Where: T s It indicates the time required for the relative spacing between trains to converge and the convoy to reach stability.

[0047] Objective 2, platoon stability: Considering the stability of adjacent train platoons, when disturbances occur, the platoon spacing error does not amplify and propagate to the vehicle ahead, which can be described as:

[0048]

[0049] Where: G d,i,i+1 (s) represents the queue spacing error transfer function, E d,i,i+1 (s), E d,i+1,i+2 (s) respectively represent e d,i,i+1 (t), e d,i+1,i+2 (t) is the Laplace transform of

[0050] When the system state satisfies equation (1), the spacing error of the system converges to zero in a finite time, and the speed of each follower train can track the speed of the leader train, achieving local stability. When the queue spacing error transfer function satisfies equation (2), the queue spacing error does not amplify and propagate to the front vehicle, and the spacing error attenuates and converges, achieving the stability of the entire queue.

[0051] In the problem of virtual coupled train platooning, the train is usually regarded as a single mass point to establish a nonlinear dynamic train model. The force analysis of the single mass point train dynamic model is as follows: Figure 4 shown.

[0052] From the analysis of the single-point model, it can be seen that the unit mass force during the operation of train i mainly includes the gravity m of the train itself i g, the support force F of the line on the train i ,u i(t) represents the effective traction per unit mass of the following train i, and the basic resistance w caused by air resistance during operation i , and the additional resistance f faced by the train in special sections i wait.

[0053] The basic resistance and additional resistance in train operation are caused by different reasons. The basic resistance is the resistance of the train running along a straight track under ideal line conditions. Due to the complex influencing factors, it cannot be calculated by theoretical formulas. It is usually based on empirical formulas obtained from field tests. The basic resistance w of a unit mass train at time t is i (t) is usually a quadratic trinomial of velocity and can be expressed as:

[0054]

[0055] Where a, b, and c are Davis coefficients, which are empirical parameters fitted based on field data and related to train characteristics. Their specific values ​​are determined by the operating model and operating status.

[0056] Train additional resistance refers to the additional resistance that the train encounters due to the existence of line slopes, curves or tunnels. It is divided into slope additional resistance, curve additional resistance and tunnel additional resistance according to different line environments. The total additional resistance can be expressed as:

[0057] f i (t) = f p +f r +f s (4)

[0058] In the formula, f p is the unit ramp resistance, f r is the unit curve resistance, f s is the unit tunnel resistance.

[0059] The train operation equation based on the single-particle dynamics model can be expressed as:

[0060]

[0061] Among them, p i (t) and v i (t) represent the position (m) and speed (m / s) of train unit i at time t; γ is the rotational mass coefficient.

[0062] definition And consider adding unknown uncontrollable interference term Φ i , the high-speed train dynamics model can be described as:

[0063]

[0064] In the formula, -ρf i (t) is the line disturbance, which is related to the additional line resistance.

[0065] S2: constructing a train queue spacing error model according to the target spacing, and determining the vehicle spacing error between each train unit and the corresponding target position based on the train queue spacing error model.

[0066] For the train spacing strategy, a fixed headway is used (i.e., the position of the head car of each train is the actual position of the train) to describe the error information of the relative spacing of trains in the virtual marshaling. The spacing error is expressed as:

[0067]

[0068] Where: e d,i,j represents the error between the running distance between the i-th train and the j-th train and the expected safety distance, where i≠j; d safe is the expected safety distance; l n is the length of the nth train, where j≤n≤i.

[0069] If the trains are of the same type, i.e., have the same length l, then the above formula (1) can be expressed as:

[0070] e d,i,j =p j -p i -(ij)(d safe +l) (8)

[0071] S3: Construct a distributed controller based on the global fast terminal sliding surface and the power reaching law.

[0072] For the design of global fast terminal sliding surface:

[0073] Sliding mode control has the advantages of fast response, no need for online system identification, and simple physical implementation. However, when the state trajectory reaches the sliding surface, chattering is likely to occur. Since there is no switching term in the designed control rate of global fast terminal sliding mode control, chattering can be effectively eliminated. At the same time, global fast terminal sliding mode control (GFTSMC) can adjust the system state through parameters to quickly reach an equilibrium state within a limited time. It can also adjust the parameters to deal with the influence of uncertain factors such as unknown interference in the system, and has good robustness. The present invention adopts global fast terminal sliding mode control and constructs the sliding surface as:

[0074]

[0075] Where: x∈R n are state variables; α, β>0; p and q are positive odd numbers and satisfy p>q.

[0076] From the above formula, we can know that:

[0077]

[0078] When the system state is far away from the switching function, that is, when x is far away from the origin, the convergence time is mainly determined by -βx q / p When the system state x is close to the equilibrium state x=0, the convergence time is mainly determined by -αx, and the distance between the current state x and the sliding surface decays exponentially rapidly.

[0079] In the above formula Written in differential form dx / dt s , can be transformed into:

[0080]

[0081] Integrating the above formula yields:

[0082]

[0083] Therefore, GFTSMC enables the system to s It converges from any initial state x(0)≠0 to the equilibrium state x=0.

[0084] Since it is difficult to achieve accurate control of the single-particle model of the virtual train formation established in the previous article with traditional centralized control, the distributed control strategy can improve the accuracy of train control by controlling different vehicles. Therefore, the present invention designs a distributed global fast terminal sliding mode controller for virtual train formation to achieve the effect that all following trains stably follow the leading train. From the sliding mode surface in formula (9), the sliding mode surface for each following train can be designed as:

[0085]

[0086] Where S = [s1, s2, …, s n ] T ,s i is the sliding surface corresponding to the controller of train i; X=[x1 x2…x n ] T , where x i =e d,i,i-1 =p i-1 -p i -d safe -l i-1 , is the tracking error between train i and the expected position, which is related to the position of the preceding train and the expected safety distance; Λ1=diag(α1,α2,…,α n ), Λ2=diag(β1,β2,…,β n ) is a positive diagonal matrix, αi , β i is the corresponding sliding surface s i The adjustment factor.

[0087] Since the initial value x of the inter-vehicle error of the virtual train marshaling system is i (0) is bounded and non-zero. According to the conclusion in formula (12), the inter-vehicle distance error can be The inner convergence is 0, and the parameter α is adjusted i , β i The size of p and q can adjust the convergence time.

[0088]

[0089] The above formula can show that when the train spacing error is on the sliding surface s i When the following error can be Converges to zero, making the fleet system stable within a workshop finite time.

[0090] For the design of reaching law:

[0091] The present invention can adopt the existing power reaching law, but in order to eliminate the steady-state convergence error of the inter-vehicle distance caused by external interference, an improved power reaching law is further designed.

[0092] The present invention adopts the improved power approaching law, which is expressed as:

[0093]

[0094] In the formula, the fal function is a nonlinear combined piecewise function, which can play a smoothing effect and further weaken the chattering. The expression is:

[0095]

[0096] In the formula: k1>0, k2>0, 0<δ≤1, θ>0.

[0097] When δ=1, the above formula can be written as:

[0098]

[0099] When |s|>1, the reaching law has a high reaching speed; when |s|≤1, the reaching law becomes a fast power reaching law, ensuring that the system state reaches the sliding surface within a finite time. + or s→0 - Sometimes, there are Therefore, no matter whether the system is far away from or close to the sliding surface, the reaching law can not only reduce the chattering, but also converge quickly, so that the system state can reach the sliding surface in a limited time from any position.

[0100] Consider the Lyapunov function:

[0101]

[0102] The derivative of V is:

[0103]

[0104] When k1>0, k2>0, we have Therefore, the stability condition is satisfied and the system state can reach the equilibrium point s = 0 under this reaching law. It can converge to zero in finite time, and the finite time calculation is calculated by calculation.

[0105] Assuming the initial condition s(0)=s0>1, the convergence process of the system is divided into two stages, namely:

[0106] 1) The first stage: s(0) = s0 → s(t1) = 1. At this time, the reaching law can be expressed as:

[0107]

[0108] Integrating both sides simultaneously gives:

[0109]

[0110] Therefore, the time required for s(0)=s0→s(t1)=1 is:

[0111]

[0112] 2) The second stage: s(t1) = 1 → s(t2) = 0. At this time, the reaching law can be expressed as:

[0113]

[0114] Therefore, the time required for s(t1)=1→s(t2)=0 is:

[0115]

[0116] In summary, the system can move from any initial state s = s0 to the sliding surface s = 0 within a finite time t1 + t2. When s(0) = s0 < 0, the proof is the same as above, that is, the system can reach equilibrium from any initial position within a finite time.

[0117] Formulas (18) to (24) prove that the designed convergence law can make the queue system reach equilibrium from any initial position within a finite time. However, unlike the finite time to reach equilibrium of the aforementioned sliding surface, the designed convergence law is to ensure that when the train spacing error is not on the GFTSMC sliding surface due to external interference factors, the spacing error will also reach the sliding surface within a finite time and keep the queue system vehicle spacing error converge to zero within a finite time, thereby achieving system robustness. The designed convergence law can also avoid negative effects such as system jitter when the error variable reaches the sliding surface that may be caused by the traditional convergence law.

[0118] Therefore, the global fast terminal sliding mode control law combined with the improved power reaching law is:

[0119]

[0120] in

[0121] In formula (25), Corresponding sliding mode control; The corresponding reaching law control; the combination of the two parts constitutes the control law of the distributed controller. The sliding mode control part in formula (25) can be directly derived through the design law of the sliding mode control law.

[0122] S4: Taking the train running speed of each train unit and the inter-vehicle distance error as input, the distributed controller (in Formula 25) is used to output the finite time control law u of each train i (t).

[0123] S5: The finite time control law u of each train unit i (t) is input into the train distributed dynamics model (in Formula 6) to realize the train virtual coupling platoon control.

[0124] In this embodiment, 1) in order to solve the problem of rapid convergence of inter-vehicle distance error in virtual coupled train queues, the present invention designs a global fast terminal sliding mode control surface (GFTSMS). By adjusting the controller parameters, the inter-vehicle distance error can be placed on the sliding mode surface at the initial moment, thereby accelerating the convergence speed of the inter-vehicle distance error and maintaining global robustness. 2) In order to solve the problem that the exponential approach law is prone to produce jitters, causing frequent acceleration and deceleration of trains, and the general exponential power approach law makes it impossible for the inter-vehicle distance error of trains to converge completely, a new variable exponential power approach law is designed. When the vehicle driving process is disturbed by the external environment, resulting in the inter-vehicle distance error being unable to remain on the sliding mode surface, the designed approach law can make the inter-vehicle distance error converge to the sliding mode surface within a limited time, reduce the jitter phenomenon, and improve ride comfort.

[0125] In order to verify the effectiveness of the control method proposed in this invention for the control of the virtual coupled train queue, the control performance of the virtual coupled high-speed train queue under the two marshaling topologies of LF and PF proposed in the previous article is simulated and verified. The simulation scenarios all use 1 pilot train and 3 following trains. The 4 trains are all CRH380A 8-marshaling (2T+6M) high-speed train EMUs. The initial position of each train is set to p i =[0,-1000,-2000,-3000], (i=0,1,2,3), the train parameters are shown in Table 1.

[0126] Table 1 Train parameters

[0127] Parameter symbol / unit Simulation value Davis coefficient a 0.53 Davis coefficient b 0.014 Davis coefficient c 0.001477 Train length l / m 203 <![CDATA[d safe / m]]> 20 <![CDATA[Maximum speed v max / (km / h)]]> 380

[0128] Design the pilot train to receive the running speed time curve requirements issued by the control center dispatcher as follows Figure 5 As shown in the figure, the speed limit requirements of the line are included. The simulation simulates the operation process of the train within 3000s, and the running distance is about 850km. Due to the speed limit, the speed curve includes 2 acceleration stages, 3 constant speed stages, 1 deceleration stage and 1 braking stage.

[0129] In order to simulate the uncontrollable disturbance factors during the operation of the train, a piecewise function is used to represent the disturbance of the train, which is expressed as:

[0130]

[0131] Where D is 10, W t is random white noise interference.

[0132] The controller (GFTSMC-VEPRL) proposed in this paper is compared with the traditional sliding mode controller under the exponential reaching law (SMC-ERL) and the global fast terminal controller under the general exponential power reaching law (GFTSMC-GPRL). The exponential reaching law used by SMC-ERL is GFTSMC-GPRL adopts the general exponential power reaching law as The simulation solver is set to ode45 solver with variable step size. The controller parameters are shown in Table 2.

[0133] Table 2 Controller parameters

[0134]

[0135] LF topology simulation results:

[0136] The simulation results of the virtual coupled train platoon using the LF communication topology are shown below: Figure 6 to Figure 9 shown. Figure 6The train position time curves under three control strategies, SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL, were compared; Figure 7 The curves of the distance between the train and the preceding train under three different control strategies are compared; Figure 8 The train speed time curves under three different control strategies are compared; Fig. 9 The train speed position time curves under three different control strategies are compared.

[0137] In order to compare the controller performance more intuitively, the comparative information of the following vehicle status under the three controllers is shown in Table 3.

[0138] Table 3 LF topology train following performance

[0139]

[0140] Combination Figure 6 to Figure 9 As can be seen from Table 3, the positions of the following trains under the SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL control methods have no overlap, such as Figure 6 As shown in (a), (b) and (c) in the figure, the safety of the convoy can be guaranteed, that is, the stability within the convoy can be guaranteed. However, the position convergence speed under the SMC-ERL control method changes suddenly, which seriously affects the smoothness of train operation and passenger comfort. Figure 7 (a), (b) and (c) are the relative inter-vehicle distance change curves of each train in the fleet under the three control methods. It can be seen that under the SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL control methods, the inter-vehicle distance between trains can be stabilized near the expected safety distance value after adjustment. Figure 7 As can be seen from (a) in the figure, under the SMC-ERL control method, due to the sudden change in the position convergence speed, the train spacing quickly stabilizes within the expected safe spacing range, which does not conform to the actual situation. Figure 7 (b) and Figure 7 In (c), we can see that under the action of GFTSMC-GPRL and GFTSMC-VEPRL, the train spacing did not quickly stabilize within the expected safe spacing range. Moreover, due to the smoother processing effect of the designed new power approach law, at 1350s when the interference disappears, the vehicle spacing jitter under the action of GFTSMC-VEPRL is much smaller than that under the action of GFTSMC-GPRL. Figure 8It can be seen from (a), (b) and (c) in the figure that under the SMC-ERL method, the following train in the platoon experienced a serious speed increase phenomenon in the initial operation stage, which corresponds to the aforementioned position convergence mutation, which does not conform to the actual characteristics of high-speed train operation; while under the GFTSMC-GPRL and GFTSMC-VEPRL methods, the following train did not experience a speed increase phenomenon in the initial operation stage, and the latter method had a much smaller speed jitter at 1350s when the interference disappeared than under the action of GFTSMC-GPRL, indicating that under the action of the convergence law of the latter, the impact of interference on the system is greatly reduced, thereby improving the robustness of the system. Fig. 9 It can be seen from (a), (b) and (c) in the figure that after the initial operation stage, the speed vibration amplitudes of the three following trains decrease successively under the three control methods, regardless of whether there is interference or not, and the riding comfort becomes better and better.

[0141] PF topology simulation results:

[0142] The simulation results of the virtual coupled train platoon using the PF communication topology are shown below: Figure 10 to Figure 13 shown. Fig.10 The train position time curves under three control strategies, SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL, were compared; Fig.11 The curves of the distance between the train and the preceding train under three different control strategies are compared; Fig.12 The train speed time curves under three different control strategies are compared; Fig.13 The train speed position time curves under three different control strategies are compared.

[0143] In order to compare the controller performance more intuitively, the comparative information of the following vehicle status under the three controllers is shown in Table 4.

[0144] Table 4 PF topology train following performance

[0145]

[0146] Combination Figure 10 to Figure 13 From Table 4, we can see that

[0147] There is no overlap in the positions of the following trains under the SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL control methods, such as Fig.10 As shown in (a), (b) and (c) in the figure, the safety of the convoy can be guaranteed, that is, the stability within the convoy can be guaranteed. However, the position convergence speed under the SMC-ERL control method changes suddenly, which seriously affects the smoothness of train operation and passenger comfort. Fig.11(a), (b) and (c) are the relative inter-vehicle distance change curves of each train in the fleet under the three control methods. It can be seen that under the SMC-ERL, GFTSMC-GPRL and GFTSMC-VEPRL control methods, the inter-vehicle distance between trains can be stabilized near the expected safety distance value after adjustment. Fig.11 As can be seen from (a) in the figure, under the SMC-ERL control method, due to the sudden change in the position convergence speed, the train spacing quickly stabilizes within the expected safe spacing range, which does not conform to the actual situation. Fig.11 (b) and Fig.11 In (c), we can see that under the action of GFTSMC-GPRL and GFTSMC-VEPRL, the train spacing did not quickly stabilize within the expected safe spacing range. Moreover, due to the smoother processing effect of the designed new power approach law, at 1350s when the interference disappears, the vehicle spacing jitter under the action of GFTSMC-VEPRL is much smaller than that under the action of GFTSMC-GPRL. Fig.12 It can be seen from (a), (b) and (c) in the figure that under the SMC-ERL method, the following train in the platoon experienced a serious speed increase phenomenon in the initial operation stage, which corresponds to the aforementioned position convergence mutation, which does not conform to the actual characteristics of high-speed train operation; while under the GFTSMC-GPRL and GFTSMC-VEPRL methods, the following train did not experience a speed increase phenomenon in the initial operation stage, and the latter method had a much smaller speed jitter at 1350s when the interference disappeared than under the action of GFTSMC-GPRL, indicating that under the action of the convergence law of the latter, the impact of interference on the system is greatly reduced, thereby improving the robustness of the system. Fig.13 It can be seen from (a), (b) and (c) in the figure that after the initial operation stage, the speed vibration amplitudes of the three following trains decrease successively under the three control methods, regardless of whether there is interference or not, and the riding comfort becomes better and better.

[0148] It is worth noting that in the PF topology, each vehicle can only obtain information about the vehicle in front, and the following vehicle cannot act in time when the running state of the vehicle in front changes. Therefore, there is a small tracking lag compared to the PF topology. Figure 8 and Fig.13 It can be seen that compared with the LF topology, the jitter of the three trains following the train speed in the PF topology increases successively, but the lag range is still within an acceptable range.

[0149] The simulation results show that for different communication topologies, the relative inter-vehicle distance jitter under the GFTSMC-VEPRL finite-time controller is smaller than that of SMC-ERL and GFTSMC-GPRL. Although the three control methods can ultimately keep the convoy at a stable following speed, under two different topologies, the jitter of the SMC-ERL and GFTSMC-GPRL methods in vehicle speed is greater than that of the GFTSMC-VEPRL method, which will reduce the smoothness and comfort of the train during operation. In addition, the position and speed of the SMC-ERL under the two topologies have serious sharp increases and mutations, which do not conform to the actual train and endanger driving safety. In comparison, the finite-time controller designed by the present invention significantly reduces the controller output jitter phenomenon in the inter-vehicle distance convergence process, and no sharp increases and mutations occur, effectively ensuring driving safety.

[0150] In the present invention, in order to solve the problem of longitudinal cooperative control and rapid error convergence of virtual coupled queues of high-speed trains, the convergence performance of the spacing of the virtual coupled queue system of trains and the stability of strong queues are improved. In view of the problem of virtual coupling control of trains, the operation mechanism of virtual coupling of high-speed trains is considered, and a single-particle dynamic model of high-speed trains, two communication topologies of pilot following (LF) and front following (PF), and a fixed headway strategy are established. A global fast terminal sliding mode surface (GFTSMS) containing train spacing errors is constructed to accelerate the convergence speed of the spacing error of coupled queue trains and achieve rapid tracking of the target speed. A new variable exponential power reaching law (VEPRL) is introduced to eliminate the steady-state convergence error of the spacing between trains caused by external interference. A global fast terminal sliding mode controller (GFTSMC-VEPRL) based on GFTSMS and VEPRL is constructed, and the finite-time stability of the coupled train queue and the rapid convergence of the train spacing error are analyzed by constructing the Lyapunov function.

[0151] The research results show that under the LF communication topology, compared with the traditional sliding mode controller (SMC-ERL) under the exponential reaching law, the global fast terminal controller (GFTSMC-GPRL) under the general exponential power reaching law and the GFTSMC-VEPRL control method proposed in the present invention, the error peak value is reduced. Under the PF communication topology of the leading vehicle following, compared with the traditional sliding mode controller (SMC-ERL) under the exponential reaching law and the global fast terminal controller (GFTSMC-GPRL) under the general exponential power reaching law, the GFTSMC-VEPRL control method proposed in the present invention also reduces the error peak value. Although the three control methods can ultimately keep the fleet at a stable following speed, under the two different topologies, the vehicle speed jitter under the GFTSMC-VEPRL method is smaller than that of the SMC-ERL and GFTSMC-GPRL methods, which will improve the smoothness and comfort of the train during operation. Therefore, it is concluded that the control strategy proposed in the present invention can effectively improve the rapid convergence performance and robust performance of the train spacing error under two different communication topologies, and improve the jitter amplitude of the train speed caused by the jitter generated by the controller, providing a reference for further enriching the theory of high-speed train automatic driving.

[0152] Example 2

[0153] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of a finite time control method for a train virtual coupling queue in Embodiment 1.

[0154] Example 3

[0155] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a train virtual coupling queue finite time control method in embodiment 1.

[0156] Example 4

[0157] A computer program product includes a computer program, which, when executed by a processor, implements the steps of a train virtual coupling queue finite time control method in embodiment 1.

[0158] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0159] The present invention uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A train virtual coupling queue finite time control method, characterized in that: The train virtual coupling queue finite time control method comprises: Construct a distributed dynamics model of the train; Constructing a train queue spacing error model according to the target spacing, and determining the vehicle spacing error between each train unit and the corresponding target position based on the train queue spacing error model; Construct a distributed controller based on global fast terminal sliding surface and power reaching law; Taking the train running speed of each train unit and the inter-vehicle distance error as input, the distributed controller is used to output a finite time control law of each train; The finite-time control law of each train unit is input into the train distributed dynamics model to realize the finite-time control of the train virtual coupling queue.

2. The train virtual coupling queue finite time control method according to claim 1 is characterized in that: The expression of the train distributed dynamics model is: In the formula, represents the position p of the i-th train unit at time t i The first derivative of (t); v i (t) represents the speed of the i-th train unit at time t; Indicates v i The first derivative of (t); g represents the acceleration of gravity; γ is the rotational mass coefficient; Φ i represents the interference term; -ρf i (t) is the line disturbance, which is related to the line additional resistance; a, b, c are Davis coefficients; u i (t) represents the control law of the i-th train unit at time t.

3. The train virtual coupling queue finite time control method according to claim 1, characterized in that: The expression of the train queue spacing error model is: In the formula, e d,i,j represents the error between the train running distance between the i-th train unit and the j-th train unit and the expected safety distance, where i≠j; d safe is the expected safety distance; l n is the length of the nth train unit, where j≤n≤i.

4. The train virtual coupling queue finite time control method according to claim 1, characterized in that: Before executing step "Building a distributed controller based on global fast terminal sliding surface and power reaching law", it also includes: An improved power reaching law is designed to construct the distributed controller based on the improved power reaching law; the expression of the improved power reaching law is: in, represents the first derivative of the global fast terminal sliding surface s; k1, k2, δ, θ represent coefficients, and k1>0, k2>0, 0<δ≤1, θ>0; the fal function is a nonlinear combined piecewise function, and the function expression is:

5. The train virtual coupling queue finite time control method according to claim 4 is characterized in that: The expression of the distributed controller is: In the formula, represents the second-order derivative of the expected position of the i-th train unit at time t, which is related to the position of the preceding vehicle and the expected safety distance; ω i represents the basic resistance received by the i-th train unit at time t; x i Represents the inter-vehicle distance error between each train unit and the corresponding target position; Represents x i The first derivative of; p and q are positive odd numbers and satisfy p>q; α i and β i is the corresponding sliding surface s i The adjustment factor.

6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the train virtual coupled queue finite time control method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the train virtual coupling queue finite time control method described in any one of claims 1 to 5 are implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the train virtual coupling queue finite time control method described in any one of claims 1 to 5 are implemented.