A high-speed train data-driven terminal sliding mode decoupling control method and device

Abstract

The present application discloses a data-driven terminal sliding mode decoupling control method and device for a high-speed train, which relates to the technical field of automatic driving of high-speed trains. The method comprises: establishing a distributed mathematical model of multiple power units of a high-speed train system, converting it into a dynamic linearized data model, splitting the parameter matrix in the dynamic linearized data model, and obtaining a decoupled dynamic linearized data model; designing a parameter matrix estimation algorithm for estimating the time-varying parameter matrix in the decoupled dynamic linearized data model, and designing an adaptive extended state observer for estimating uncertain terms; designing a nonlinear sliding mode function and a hyperbolic sliding mode convergence law; deriving a data-driven terminal sliding mode decoupling control method, and obtaining the control force and speed results of each power unit of the high-speed train system. The present application can alleviate the system control force chattering phenomenon, and has fast convergence and strong anti-interference ability.

Description

Technical Field

[0001] The present application relates to the technical field of high-speed train automatic driving, and in particular to a method and device for high-speed train data-driven terminal sliding mode decoupling control. Background Art

[0002] For high-speed trains, achieving high-performance, stable tracking control for automated driving systems under unknown disturbances is an urgent need. However, current tracking control methods for automated driving systems often suffer from chattering, making it difficult to meet the high-stability tracking control requirements of actual high-speed train operations. Summary of the Invention

[0003] The purpose of this application is to provide a high-speed train data-driven terminal sliding mode decoupling control method and device, which can alleviate the vibration phenomenon and has fast convergence and strong anti-interference ability.

[0004] To achieve the above objectives, this application provides the following solutions.

[0005] In a first aspect, the present application provides a high-speed train data-driven terminal sliding mode decoupling control method, comprising:

[0006] Based on the coupling force between the power units of the high-speed train system, a distributed mathematical model of multiple power units of the high-speed train system is established;

[0007] The distributed mathematical model of multiple power units with input constraints is converted into a virtual data model to achieve dynamic linear transformation of the nonlinear system and obtain a dynamic linear data model.

[0008] The parameter matrix in the dynamic linearization data model is split to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model is composed of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance includes uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix;

[0009] Designing a parameter matrix estimation algorithm and an adaptive extended state observer; the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearization data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance;

[0010] Based on the dynamic linearized data model, parameter matrix estimation algorithm and adaptive extended state observer, a nonlinear sliding mode function and a hyperbolic sliding mode reaching law are designed respectively;

[0011] Based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introducing the control force saturation mechanism, a data-driven terminal sliding mode decoupling control method is derived, and the control force and speed results of each power unit of the high-speed train system are obtained.

[0012] In a second aspect, the present application provides a high-speed train data-driven terminal sliding mode decoupling control device, comprising:

[0013] A multi-power unit distributed mathematical model establishment module, configured to establish a multi-power unit distributed mathematical model of the high-speed train system based on the coupling forces between the power units of the high-speed train system;

[0014] The dynamic linearization data model conversion module is used to convert the distributed mathematical model of multiple power units with input constraints into a virtual data model, realize the dynamic linearization conversion of the nonlinear system, and obtain the dynamic linearization data model;

[0015] A parameter matrix splitting module is used to split the parameter matrix in the dynamic linearization data model to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model consists of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance consists of uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix;

[0016] A parameter estimation module is used to design a parameter matrix estimation algorithm and an adaptive extended state observer; the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearization data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance;

[0017] A sliding mode function and reaching law design module, for designing a nonlinear sliding mode function and a hyperbolic sliding mode reaching law, respectively, based on the dynamic linearization data model, the parameter matrix estimation algorithm, and the adaptive extended state observer;

[0018] The data-driven terminal sliding mode decoupling control module is used to derive the data-driven terminal sliding mode decoupling control method based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introduce the control force saturation mechanism to obtain the control force and speed results of each power unit of the high-speed train system.

[0019] In a third aspect, the present application provides 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 above-mentioned high-speed train data-driven terminal sliding mode decoupling control method.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned high-speed train data-driven terminal sliding mode decoupling control method.

[0021] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0022] The present application provides a data-driven terminal sliding mode decoupling control method and device for a high-speed train. By decoupling a dynamic linearized data model, a parameter matrix estimation algorithm, an adaptive extended state observer, a nonlinear sliding mode function, and a hyperbolic sliding mode convergence law, and introducing a control force saturation mechanism, a data-driven terminal sliding mode decoupling control method is derived. The designed nonlinear sliding mode function and hyperbolic sliding mode convergence law simultaneously meet the control requirements of fast convergence and vibration elimination, and have fast convergence and strong anti-interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is an application environment diagram of a high-speed train data-driven terminal sliding mode decoupling control method in one embodiment of the present application.

[0025] Figure 2 A flow chart of a high-speed train data-driven terminal sliding mode decoupling control method provided in one embodiment of the present application.

[0026] Figure 3 A schematic diagram of the dynamic analysis of the high-speed train operation process provided in one embodiment of the present application.

[0027] Figure 4 This is a control block diagram of the high-speed train data-driven terminal sliding mode decoupling control (DDTSMDC) provided in one embodiment of the present application.

[0028] Figure 5 This is a schematic diagram of the speed tracking curves of each power unit of a high-speed train under the DDTSMDC scheme provided in one embodiment of the present application.

[0029] Figure 6A schematic diagram of the speed tracking curves of each power unit of a high-speed train under the PID scheme provided in one embodiment of the present application.

[0030] Figure 7 A schematic diagram of a speed error curve of a high-speed train power unit under the DDTSMDC scheme provided in one embodiment of the present application.

[0031] Figure 8 This is a schematic diagram of the second speed error curve of the high-speed train power unit under the DDTSMDC scheme provided in one embodiment of the present application.

[0032] Figure 9 This is a schematic diagram of the three-speed error curves of the high-speed train power unit under the DDTSMDC scheme provided in one embodiment of the present application.

[0033] Figure 10 A schematic diagram of a speed error curve of a high-speed train power unit under the PID scheme provided in one embodiment of the present application.

[0034] Figure 11 This is a schematic diagram of the second speed error curve of the high-speed train power unit under the PID scheme provided in one embodiment of the present application.

[0035] Figure 12 This is a schematic diagram of the three-speed error curves of the high-speed train power unit under the PID scheme provided in one embodiment of the present application.

[0036] Figure 13 This is a schematic diagram of the acceleration curves of each power unit of a high-speed train under the DDTSMDC scheme provided in one embodiment of the present application.

[0037] Figure 14 This is a schematic diagram of the acceleration curves of each power unit of a high-speed train under the PID scheme provided in one embodiment of the present application.

[0038] Figure 15 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0040] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0041] The high-speed train data driven terminal sliding mode decoupling control method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the tracking control request to be processed to the server 104. After the server 104 receives the tracking control request to be processed, it obtains the control force and speed results of each power unit of the high-speed train system based on the data-driven terminal sliding mode decoupling control method. The server 104 can feed back the obtained control force and speed results of each power unit of the high-speed train system to the terminal 102. In addition, in some embodiments, the high-speed train data-driven terminal sliding mode decoupling control method can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly perform train tracking control for the tracking control request to be processed, or the server 104 can obtain the tracking control request to be processed from the data storage system and perform train tracking control for the tracking control request to be processed.

[0042] Terminal 102 may include, but is not limited to, various desktop computers, laptops, smartphones, tablet computers, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Server 104 may be implemented as a standalone server or a server cluster consisting of multiple servers, or may be a cloud server.

[0043] In an exemplary embodiment, Figure 2 As shown, a high-speed train data driven terminal sliding mode decoupling control method is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used for explanation, including the following steps 201 to 206.

[0044] Step 201 : establishing a multi-power unit distributed mathematical model of the high-speed train system based on the coupling force between the power units of the high-speed train system.

[0045] In step 202 , the distributed mathematical model of multiple power units with input constraints is converted into a virtual data model to achieve dynamic linearization conversion of the nonlinear system, thereby obtaining a dynamic linearization data model.

[0046] Step 203: Split the parameter matrix in the dynamic linearization data model to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model consists of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance consists of uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix.

[0047] Step 204 , designing a parameter matrix estimation algorithm and an adaptive extended state observer (AESO); the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearized data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance.

[0048] Step 205 : Based on the dynamic linearized data model, the parameter matrix estimation algorithm, and the adaptive extended state observer, a nonlinear sliding mode function and a hyperbolic sliding mode reaching law are designed respectively.

[0049] Step 206, based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introducing the control force saturation mechanism, derives the data-driven terminal sliding mode decoupling control method to obtain the control force and speed results of each power unit of the high-speed train system.

[0050] By implementing steps 201 through 206, a data-driven terminal sliding mode decoupling control method is derived by decoupling a dynamic linearized data model, employing a parameter matrix estimation algorithm, an adaptive extended state observer, a nonlinear sliding mode function, and a hyperbolic sliding mode reaching law, and introducing a control force saturation mechanism. The designed nonlinear sliding mode function and hyperbolic sliding mode reaching law simultaneously achieve the control requirements of rapid convergence and chattering elimination, exhibiting rapid convergence and strong disturbance rejection. To achieve high-performance, stable tracking control of a high-speed train automatic driving system under unknown disturbances, a multi-input, multi-output (MIMO) data-driven variable structure control scheme with chattering mitigation is proposed. This is a data-driven approach based on extended state observer decoupling.

[0051] (1) In another exemplary embodiment of the present application, Figure 3As shown in the figure, firstly, the dynamic analysis of the high-speed train operation process is carried out to establish a distributed mathematical model of multiple power units of the high-speed train. The power units of the high-speed train are connected together by a coupler device. During the operation of the high-speed train, there will be mutual coupling forces between the power units, so the entire high-speed train can be regarded as a strongly coupled system composed of multiple power units. During the operation of the high-speed train, the high-speed train is subjected to the traction / braking force generated by the traction transmission device, as well as the resistance generated by environmental factors and route factors. The force analysis of the high-speed train multi-power unit model is shown in the figure. Figure 3 As shown. During the operation of a high-speed train, each power unit provides a certain amount of traction, and the extension and retraction of the coupler will cause relative movement between adjacent carriages. The main forces acting on each power unit include: control force, vehicle interaction force, air resistance, and additional resistance. Based on the analysis, the dynamic model of the high-speed train shown in the following formula (1) can be established, that is, the distributed mathematical model of multiple power units is expressed as follows:

[0052] (1);

[0053] in, Indicates power unit speed; is the current sampling time; For the power unit The acceleration coefficient; Indicates power unit the net force applied; For the power unit control power; For the power unit The basic resistance encountered; For the power unit and power unit Workshop forces between For the power unit and power unit +1 workshop force; , is the total number of power units.

[0054] The specific forms of basic resistance and workshop force are as follows:

[0055] (2);

[0056] Where: Indicates power unit Displacement; Parameters is the basic drag coefficient, which has a high degree of uncertainty; is the elastic coefficient of the adjacent power unit; For the power unit +1 displacement; is the damping coefficient, For the power unit +1 for speed.

[0057] Applying the forward Euler formula to the distributed mathematical model of multiple power units of the high-speed train system, that is, formula (1), the following discrete-time nonlinear model can be obtained:

[0058] (3);

[0059] Where: for Moment power unit +1 speed; is a nonlinear time-varying function; They are Timing, power unit -1. Power unit , power unit +1 speed; They are Timing, power unit -1. Power unit , power unit +1 displacement, for Moment power unit control power.

[0060] (2) The distributed mathematical model of multiple power units with input constraints is equivalent to a virtual data model to realize the dynamic linearization conversion of the nonlinear system.

[0061] Convert all displacement terms in formula (3) into forms related to speed and control force, and express external disturbances, parameter estimation errors, etc. as total disturbance terms. Then formula (3) can be equivalent to the following MIMO high-speed train discrete-time nonlinear system:

[0062] (4);

[0063] in, for The total speed of the high-speed train system at that moment; , are the total speed and total control force of the high-speed train system respectively; They are At this moment, the first power unit and the The speed of each power unit; They are At this moment, the first power unit and the Control force of each power unit; is the bounded unknown disturbance of the high-speed train system; They are At this moment, the first power unit and the The bounded unknown disturbance of the power unit; let , is a constant; is a nonlinear function; 、 、 are the orders of the system output (i.e. speed), input (control force) and unknown term (also called uncertain term), respectively.

[0064] Considering that the output torque of the traction motor during the actual operation of a high-speed train is limited by the motor power, the system control force and the rate of change of the control force are subject to the input constraints shown in the following equations:

[0065] (5);

[0066] Where: and Respectively represent the maximum braking force and maximum traction force of the power unit; is a constant representing the constraint of any control force; for -1 moment power unit control power, is the rate of change of control force.

[0067] The following two assumptions are made for the MIMO high-speed train discrete-time nonlinear system subject to input constraint (5).

[0068] Assumption 1: Nonlinear time-varying function The partial derivative of the control force exists, is not zero and has the same sign.

[0069] Assumption 2: The system satisfies the generalized Lipschitz condition, namely:

[0070] (6);

[0071] Where: ; and Represent nonlinear functions Middle elements and the Lipschitz constant, , is a nonlinear function The total number of elements in .

[0072] Theorem 1: For a discrete-time high-speed train system (4) that satisfies Assumptions 1-2 and is subject to input constraints (5), when When , there must be a time-varying parameter matrix , so that the discrete-time high-speed train system (4) can be transformed into the following virtual but equivalent dynamic linearized data model:

[0073] (7);

[0074] Where: is the total uncertainty term of the dynamic linearized data model, ; Any sampling time , the time-varying parameter matrix are bounded, for Middle Rank Elements of the column, , .

[0075] (3) Split the parameter matrix in the dynamic linearization data model into a parameter matrix consisting of a diagonal system matrix, unknown terms (also called uncertain terms), and coupling effects.

[0076] In order to further improve the dynamic performance of the system, the parameter matrix in the dynamic linearization data model is split into a second parameter matrix consisting of a diagonal system matrix, uncertain terms, and coupling effects with the help of the disturbance decoupling idea. Then, Equation (7) can be rewritten as the decoupled dynamic linearization data model shown in the following equation:

[0077] (8);

[0078] Where: for The change in the total control force of the high-speed train system at that moment; the superscript T indicates transposition; for The high-speed train system contains generalized bounded disturbances with uncertainties and system coupling effects; They are At this moment, the first power unit and the Generalized bounded disturbance of a dynamic unit; ; for The time-varying parameter matrix after moment decoupling, is the diagonal matrix symbol, is the element in the first row and first column of the decoupled time-varying parameter matrix, is the first parameter in the decoupled time-varying parameter matrix Rank Elements of a column.

[0079] (IV) A parameter matrix estimation algorithm and an adaptive extended state observer are designed to process the two matrices respectively, solving the estimation of data model parameters and total uncertainty.

[0080] Since the time-varying parameter matrix is unknown, and the parameter matrix estimation algorithm shown below is introduced to estimate the time-varying parameter matrix:

[0081] (9);

[0082] Where: for Time-varying parameter matrix The estimated value of for Middle Rank Elements of the column; for -1 time-varying parameter matrix estimated value of; is the step size factor, ; for The change in the total speed of the high-speed train system at any given moment; for The change in the total control force of the high-speed train system at time -1; the superscript T indicates transposition; is the weight factor.

[0083] In order to enhance the robustness of the parameter matrix estimation algorithm, the following parameter reset algorithm is given:

[0084] (10)

[0085] Where: for Middle Rank Elements of the column; is a small positive number; yes The initialization value of yes The initialization value of is the first time-varying parameter matrix at time 1 Rank Elements of a column.

[0086] Due to uncertainty is unknown, the AESO estimation uncertainty is designed as shown in the following formula :

[0087] (11);

[0088] Where: for +1 estimated value of the total speed of the high-speed train system; for Total speed of the high-speed train system estimated value of; for Generalized bounded perturbation of high-speed train system with time estimated value of; for Time-varying parameter matrix after moment decoupling estimated value of; for The change in the total control force of the high-speed train system at each moment; and is the adaptive extended state observer (AESO) gain; for The estimated value of the generalized bounded disturbance of the high-speed train system at time +1; the state variables and state variable observations of AESO are defined as and .

[0089] (V) Based on the split dynamic linearized data model, parameter matrix estimation algorithm and adaptive extended state observer, a nonlinear sliding mode function and a hyperbolic sliding mode convergence law are designed respectively.

[0090] Define the system speed error as , ,in is the desired output. In order to enhance the control performance of the system, a new discrete-time terminal sliding mode function is considered, that is, the nonlinear sliding mode function is as follows:

[0091] (12);

[0092] Where: for Sliding mode variables of high-speed train systems; They are At this moment, the first power unit and the Sliding mode variables of the power units; , Both are adjustable weight parameters; for System speed error of high-speed train system; for -1 time high-speed train system system speed error; is the ratio of two odd numbers; is an adjustable parameter that affects the error convergence bound of the closed-loop system.

[0093] Considering that traditional discrete sliding mode control cannot simultaneously achieve the control requirements of fast convergence and chattering elimination, an improved hyperbolic sliding mode reaching law is proposed:

[0094] (13);

[0095] (14);

[0096] Where: for +1 moment sliding mode variables of the high-speed train system; is a symbolic function; is the convergence parameter, satisfying ; and It is a parameter that adjusts the speed of change of the exponential term; is the gain parameter. and are two nonlinear functions, They are The first row and first column element in Column elements, They are The 2nd row and 1st column element in Column elements, when the sliding mode variable is large, Reduce, while Increase, thereby accelerating the system's convergence to the sliding surface. When the sliding mode variable is small, Increase, Reduced, a smaller controller gain is obtained, and the chattering phenomenon is alleviated.

[0097] (6) Based on a decoupled dynamic linearization data model, a parameter matrix estimation algorithm, an adaptive extended state observer, a nonlinear sliding mode function, and a hyperbolic sliding mode reaching law, and by introducing a control force saturation mechanism, a data-driven terminal sliding mode decoupling control (DDTSMDC) method is derived to achieve rapid error convergence and mitigate system chattering. The proposed scheme uses only input and output data from the high-speed train system throughout the entire design process, making it easy to implement and offering strong parameter adaptability and anti-interference capabilities.

[0098] In summary, the proposed DDTSMDC scheme can be obtained by combining the decoupled dynamic linearization data model, parameter matrix estimation algorithm, AESO, nonlinear sliding mode function and hyperbolic sliding mode reaching law. Its control block diagram is shown in the figure below. Figure 4 shown.

[0099] (15);

[0100] Where, is the expected trajectory, They are Moment 1 and The speed of the power unit.

[0101] (VII) Finally, the proposed scheme was compared and tested using the CRH380A high-speed train simulation test bench equipped in the laboratory. The simulation results show that the speed tracking errors of the power units of the high-speed train under the proposed control scheme are all within [-0.138 km / h, 0.142 km / h], and the control force and acceleration are within [-52.1 kN, 46.5 kN] and [-0.557 m / s 2 , 0.478 m / s 2 ] and the changes are smooth, and the system vibration level is low.

[0102] Simulation Setup: During operation, white noise was introduced to simulate the external interference encountered during actual high-speed train operation to verify the robustness of the proposed algorithm. Vehicle information and control strategies were input into the simulation testbed. Train speed, control force, position, and other information were recorded for various schemes. The schemes were then compared with the MFASMC and MFAC schemes, which are also based on a dynamic linearized data model, and the GPC scheme, which is based on a system model.

[0103] (1) The DDTSMDC scheme proposed in this application: The initial conditions of the system are set as , The controller parameters are set to , , , , , , , , , , , .

[0104] (2) MFASMC scheme: The initial conditions of the system are set to , The controller parameters are set to , , , , , , , .

[0105] (3) MFAC scheme: The initial conditions of the system are set to , The controller parameters are set to , , , , .

[0106] (4) GPC scheme: The initial conditions of the system are set to The controller parameters are set as follows: prediction time domain and control time domain are =3, =2; forgetting factor =0.95.

[0107] Simulation analysis: AESO's observations of the total system uncertainty, including disturbance estimates for power units 1, 2, and 3, show that AESO estimates disturbances well, with an error within ±0.005. This total uncertainty will be compensated for in the DDTSMDC scheme.

[0108] Figure 5-Figure 6 The speed tracking curves of each power unit of the high-speed train from Jinan West Station to Xuzhou East Station under the DDTSMDC and existing PID schemes are shown. It can be seen that the tracking results of the four schemes will not exceed the system speed limit curve, indicating that both schemes meet the most basic operation safety requirements. Figure 7 、 Figure 8 and Figure 9 They are the speed error curves of power unit 1, power unit 2 and power unit 3 from Jinan West Station to Xuzhou East Station under the DDTSMDC scheme of this application, Figure 10 、 Figure 11 and Figure 12 The speed error curves for power units 1, 2, and 3 from Jinan West Station to Xuzhou East Station under the PID scheme are shown. The tracking error range for each power unit in the DDTSMDC scheme remains stable between [-0.138 km / h and 0.147 km / h] throughout the entire process. However, due to the improved control law and AESO, the DDTSMDC scheme significantly outperforms the PID scheme in terms of rapid convergence and disturbance rejection.

[0109] This application also obtains the control force curves of each power unit of the high-speed train under the DDTSMDC, MFASMC, MFAC, and GPC schemes, showing the changes in the control force of the high-speed train under different control schemes. During the starting, braking, and inertia phases of the DDTSMDC scheme, the control force of each power unit remains stable, with the traction / braking force range being [-52.1 kN, 46.5 kN]. In addition, during the operating condition transition period, the change in control force also shows a smooth transition with a moderate rate of change. The traction / braking force range of the MFASMC scheme is within [-57.4 kN, 49.8 kN]. Due to the introduction of the terminal sliding surface and hyperbolic function, the DDTSMDC scheme significantly alleviates the system chattering phenomenon compared to the MFASMC scheme. The MFAC and GPC schemes experience large variations in control force during starting and braking, with the control force ranges being [-56.3 kN, 49.1 kN] and [-61.4 kN, 47.2 kN], respectively. Furthermore, the control force variations for both schemes are frequent and have large amplitudes. In actual control, most actuators cannot withstand the frequent switching between positive and negative outputs, posing a certain degree of safety hazard to train operation.

[0110] Figure 13 and Figure 14 The acceleration curves of each power unit of the high-speed train from Jinan West Station to Xuzhou East Station under DDTSMDC and PID schemes are shown. It is not difficult to see that the acceleration of the PID scheme changes too fast, ranging from [-0.6 m / s 2 , 0.537 m / s 2 ], the change amplitude is large. However, the acceleration transition of the high-speed train using the DDTSMDC scheme is relatively smooth, except for the starting stage, the range is [-0.558 m / s 2 , 0.478 m / s 2 ], the amplitude is smaller than that of the PID solution, while also meeting the passenger comfort requirements.

[0111] Furthermore, in order to more intuitively analyze the control performance of each controller, the following performance indicators are considered to evaluate the controller. The DDTSMDC method of this application is compared with the MFASMC, MFAC and GPC schemes, and the performance indicators are shown in Table 1. It is not difficult to see that the MSE value and IAE value of the DDTSMDC method are smaller than those of the other three methods; the maximum acceleration / deceleration reflects the stability of the system input. The maximum acceleration / deceleration of the train using the DDTSMDC method is 0.558m / s 2 , the change is small; the maximum acceleration / deceleration of MFASMC and GPC methods is larger, which is 0.586 m / s respectively. 2 and 0.613 m / s 2In addition, the energy consumption index can reflect the energy loss of the train. Calculations show that compared with the MFASMC, MFAC, and GPC schemes, the DDTSMDC scheme saves 7.7%, 4.5%, and 18.5% of energy, respectively.

[0112] Table 1 Performance indicators of DDTSMDC method, MFASMC, MFAC and GPC schemes

[0113]

[0114] The present application also provides an application scenario, which applies the above-mentioned high-speed train data-driven terminal sliding mode decoupling control method. Specifically: the high-speed train data-driven terminal sliding mode decoupling control method provided in this embodiment can be applied in the high-speed train automatic driving scenario. The high-speed train automatic driving scenario includes a request sending link and a data-driven terminal sliding mode decoupling control link; the control request to be processed enters the data-driven terminal sliding mode decoupling control link from the request sending link to obtain the input and output results of the corresponding high-speed train system. The high-speed train data-driven terminal sliding mode decoupling control method provided in this embodiment belongs to the data-driven terminal sliding mode decoupling control link. Specifically, in the process of the data-driven terminal sliding mode decoupling control link for the control request to be processed, the control force and speed results of each power unit of the high-speed train system can be obtained based on the data-driven terminal sliding mode decoupling control method.

[0115] Based on the same inventive concept, embodiments of the present application also provide a high-speed train data-driven terminal sliding mode decoupling control device for implementing the aforementioned high-speed train data-driven terminal sliding mode decoupling control method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the high-speed train data-driven terminal sliding mode decoupling control device provided below can be found in the above-mentioned limitations of the high-speed train data-driven terminal sliding mode decoupling control method and will not be further elaborated here.

[0116] In an exemplary embodiment, a high-speed train data-driven terminal sliding mode decoupling control device is provided, including the following modules.

[0117] The multi-power unit distributed mathematical model establishment module is used to establish the multi-power unit distributed mathematical model of the high-speed train system based on the coupling force between the power units of the high-speed train system.

[0118] The dynamic linearization data model conversion module is used to convert the distributed mathematical model of multiple power units with input constraints into a virtual data model, realize the dynamic linearization conversion of the nonlinear system, and obtain the dynamic linearization data model.

[0119] A parameter matrix splitting module is used to split the parameter matrix in the dynamic linearization data model to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model consists of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance consists of uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix.

[0120] The parameter estimation module is used to design a parameter matrix estimation algorithm and an adaptive extended state observer; the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearization data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance.

[0121] The sliding mode function and reaching law design module is used to design a nonlinear sliding mode function and a hyperbolic sliding mode reaching law based on the dynamic linearization data model, the parameter matrix estimation algorithm and the adaptive extended state observer.

[0122] The data-driven terminal sliding mode decoupling control module is used to derive the data-driven terminal sliding mode decoupling control method based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introduce the control force saturation mechanism to obtain the control force and speed results of each power unit of the high-speed train system.

[0123] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 15 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store high-speed train control data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a high-speed train data-driven terminal sliding mode decoupling control method is implemented.

[0124] Those skilled in the art will understand that Figure 15The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0125] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0126] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0127] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0128] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0129] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0130] The technical features of the above embodiments can 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.

[0131] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A high-speed train data-driven terminal sliding mode decoupling control method, characterized in that: The high-speed train data-driven terminal sliding mode decoupling control method includes: Based on the coupling force between the power units of the high-speed train system, a multi-power unit distributed mathematical model of the high-speed train system is established; the multi-power unit distributed mathematical model is expressed as follows: ； in, Indicates power unit speed; is the current sampling time; For the power unit The acceleration coefficient; Indicates power unit the net force applied; For the power unit control power; For the power unit The basic resistance encountered; For the power unit and power unit Workshop forces between For the power unit and power unit +1 workshop force; , is the total number of power units; The distributed mathematical model of multiple power units with input constraints is converted into a virtual data model to achieve dynamic linear transformation of the nonlinear system and obtain a dynamic linear data model. The parameter matrix in the dynamic linearization data model is split to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model consists of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance consists of uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix; the decoupled dynamic linearization data model is expressed as follows: ； in, for The total speed of the high-speed train system at that time; for The total speed of the high-speed train system at that time; They are At this moment, the first power unit and the The speed of each power unit; for The total control force of the high-speed train system at any time, for The change in the total control force of the high-speed train system at each moment; They are At this moment, the first power unit and the The control force of a power unit; the superscript T indicates transposition; for Generalized bounded perturbations of high-speed train systems at different times; They are At this moment, the first power unit and the The generalized bounded perturbation of a dynamic unit, for The time-varying parameter matrix after moment decoupling, is the diagonal matrix symbol, is the element in the first row and first column of the decoupled time-varying parameter matrix, is the first parameter in the decoupled time-varying parameter matrix Rank Elements of the column; Designing a parameter matrix estimation algorithm and an adaptive extended state observer; the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearization data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance; Based on the dynamic linearized data model, parameter matrix estimation algorithm and adaptive extended state observer, a nonlinear sliding mode function and a hyperbolic sliding mode reaching law are designed respectively; Based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introducing the control force saturation mechanism, a data-driven terminal sliding mode decoupling control method is derived, and the control force and speed results of each power unit of the high-speed train system are obtained.

2. The high-speed train data-driven terminal sliding mode decoupling control method according to claim 1, characterized in that: The calculation formula of the parameter matrix estimation algorithm to estimate the time-varying parameter matrix is ​​expressed as follows: ； Where: for Time-varying parameter matrix estimated value of; for -1 time-varying parameter matrix estimated value of; is the step size factor; for The change in the total speed of the high-speed train system at any given moment; for The change in the total control force of the high-speed train system at time -1; the superscript T indicates transposition; is the weight factor.

3. The high-speed train data-driven terminal sliding mode decoupling control method according to claim 1, characterized in that: The calculation formula of the uncertainty estimation of the adaptive extended state observer is expressed as follows: ； Where: for +1 estimated value of the total speed of the high-speed train system; for Total speed of the high-speed train system estimated value of; for Generalized bounded perturbation of high-speed train system with time estimated value of; for Time-varying parameter matrix after moment decoupling estimated value of; for The change in the total control force of the high-speed train system at each moment; and is the adaptive extended state observer gain; for Estimates of the generalized bounded perturbations of the high-speed train system at time +1.

4. The high-speed train data-driven terminal sliding mode decoupling control method according to claim 1, characterized in that: The nonlinear sliding mode function is expressed as follows: ； Where: for Sliding mode variables of high-speed train systems; They are At this moment, the first power unit and the Sliding mode variables of the power units; , Both are adjustable weight parameters; for System speed error of high-speed train system; for -1 time high-speed train system system speed error; is the ratio of two odd numbers; Is a tunable parameter.

5. The high-speed train data-driven terminal sliding mode decoupling control method according to claim 1, characterized in that: The hyperbolic sliding mode reaching law is expressed as follows: ； in, for +1 moment sliding mode variables of the high-speed train system; and are two nonlinear functions, for The sliding mode variables of the high-speed train system at that time, is a symbolic function.

6. A high-speed train data-driven terminal sliding mode decoupling control device, characterized in that: For implementing the high-speed train data-driven terminal sliding mode decoupling control method according to claim 1, the high-speed train data-driven terminal sliding mode decoupling control device comprises: A multi-power unit distributed mathematical model establishment module, configured to establish a multi-power unit distributed mathematical model of the high-speed train system based on the coupling forces between the power units of the high-speed train system; The dynamic linearization data model conversion module is used to convert the distributed mathematical model of multiple power units with input constraints into a virtual data model, realize the dynamic linearization conversion of the nonlinear system, and obtain the dynamic linearization data model; A parameter matrix splitting module is used to split the parameter matrix in the dynamic linearization data model to obtain a decoupled dynamic linearization data model; the decoupled dynamic linearization data model consists of a decoupled time-varying parameter matrix and a generalized bounded disturbance; the generalized bounded disturbance consists of uncertain terms and system coupling effects; the parameter matrix in the dynamic linearization data model includes a time-varying parameter matrix; A parameter estimation module is used to design a parameter matrix estimation algorithm and an adaptive extended state observer; the parameter matrix estimation algorithm is used to estimate the time-varying parameter matrix in the decoupled dynamic linearization data model, and the adaptive extended state observer is used to estimate the generalized bounded disturbance; A sliding mode function and reaching law design module, for designing a nonlinear sliding mode function and a hyperbolic sliding mode reaching law, respectively, based on the dynamic linearization data model, the parameter matrix estimation algorithm, and the adaptive extended state observer; The data-driven terminal sliding mode decoupling control module is used to derive the data-driven terminal sliding mode decoupling control method based on the decoupled dynamic linearization data model, parameter matrix estimation algorithm, adaptive extended state observer, nonlinear sliding mode function and hyperbolic sliding mode reaching law, and introduce the control force saturation mechanism to obtain the control force and speed results of each power unit of the high-speed train system.

7. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the high-speed train data-driven terminal sliding mode decoupling control method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the high-speed train data-driven terminal sliding mode decoupling control method according to any one of claims 1 to 5 is implemented.

Images