Freight train speed control method, electronic equipment and medium

Abstract

The invention provides a freight train speed control method, an electronic device and a medium, and the method comprises the steps: building a freight train simple substance point time-delay model, predicting the control quantity of a freight train after a certain time delay through combining the current control quantity and a target speed, eliminating the transmission delay of the control quantity in a freight train system, and improving the speed of the freight train. The purpose of reducing overshoot of a speed tracking curve is achieved; by adding observation on a total disturbance differential value in the expansion observer, the speed tracking and control precision is improved, and the accurate estimation on the total disturbance of the freight train when the freight train runs on a complex line is improved; the anti-slip control of the freight train is realized by combining the creep rate observed by the unscented Kalman filter, the slip phenomenon of the freight train is prevented, the anti-interference capability of the freight train is improved, the tracking and anti-slip control of the target speed are realized, and the speed control and operation stability of the freight train is improved.

Description

Technical Field

[0001] The invention belongs to the field of vehicle automatic control, and in particular relates to a freight train speed control method, electronic equipment and medium. Background Art

[0002] Freight trains play an important role in transportation with their large capacity, economic efficiency, environmental friendliness, safety and reliability. Freight trains are complex electromechanical hysteresis systems with high nonlinearity, large inertia, harsh operating environment and high control difficulty. The operation process is a complex system closely coupled with a complex environment, forming a more complex nonlinear, multi-constrained, multi-objective and time-varying process, which will bring a long control delay to the traction / braking system of the freight train, causing the tracking speed curve of the freight train to have a large overshoot at the speed switching point. How to solve the delay problem in the control system is of great significance to the real-time tracking of the target speed and precise parking of the freight train.

[0003] At present, there are few speed control methods for freight trains with large capacity, large inertia, and pure lag, especially the tracking curve overshoot caused by time lag is difficult to solve. At the same time, in the case of idling slip, due to the characteristics of large inertia and pure lag, the longitudinal speed of freight trains will have a large tracking error with the target speed.

[0004] The ADRC method developed in recent years has been widely used in trains, solving the problems of poor anti-interference ability, low tracking accuracy, high computational cost and strong model dependence in the classical control method. However, the existence of delay still leads to large overshoot; wheel slippage or idling during train speed tracking causes the ADRC to be unstable.

[0005] Patent application CN112346346A discloses a heavy-load train speed tracking control method, which includes establishing a heavy-load train multi-particle unit displacement model, using a robust adaptive self-disturbance rejection control method, minimizing the error between the actual running speed of the train and the target speed, improving the anti-disturbance capability of the heavy-load train and achieving the tracking control of the train to the target speed. It does not consider the overshoot problem of the tracking curve caused by time lag, and the speed tracking control accuracy is low. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies in the prior art and provide a freight train speed tracking and anti-creep method, electronic equipment and medium, which solves the problem of tracking curve overshoot caused by time lag and improves the accuracy of freight train speed control.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A method for controlling the speed of a freight train includes the following implementation process:

[0009] Get the target speed of the freight train;

[0010] A single-mass-point time-delay model of a freight train is established, and the single-mass-point time-delay model is shown in the following formula:

[0011]

[0012] A tracking differentiator is used to obtain the transition process and velocity derivative of the target velocity; a high-order extended state observer is used to determine a velocity estimate, an acceleration estimate, a total disturbance estimate, and a differential value of the total disturbance estimate;

[0013] Calculating an error between the transition process and the velocity estimate as a first error; calculating an error between the velocity derivative and the acceleration estimate as a second error;

[0014] Using a nonlinear error feedback device to design a control law for the first error and the second error to obtain a virtual control amount;

[0015] According to the virtual control amount, the total disturbance estimation value and the differential value of the total disturbance estimation value, the actual control amount is obtained; the expression of the actual control amount is as follows:

[0016]

[0017] A hyperspectral camera is used to identify the track contact conditions of freight trains in real time, select the anti-slip coefficient, calculate the creep rate and adhesion coefficient, and obtain the creep characteristic curve.

[0018] The unscented Kalman observer is used to observe the speed and creep rate of the freight train in real time, and the adhesion coefficient under the current operating conditions of the freight train is obtained according to the creep characteristic curve; the anti-creep threshold is calculated by the adhesion coefficient, and the anti-creep threshold u y The expression is as follows:

[0019] u y =μgcosθ-f / m

[0020] If the actual control amount is less than or equal to the anti-creep threshold, the freight train runs according to the actual control amount; if the actual control amount is greater than the anti-creep threshold, the freight train runs according to the anti-creep threshold;

[0021] Among them, f(x,v) is the high-order ADRC control system function, x is the real-time position of the freight train, v is the target speed, is the acceleration of the freight train, u(k-τ) is the control quantity after the delay τ, τ is the delay of the freight train, d(k) is the total disturbance, k is time, b is the compensation of the high-order active disturbance rejection control, y is the output speed of the freight train traction braking system; v1 is the transition process of the target speed, v2 is the speed derivative, z1 is the speed estimate, z2 is the acceleration estimate, e1 is the first error, e2 is the second error, z3 is the total disturbance estimate, z4 is the differential value of the total disturbance estimate, k p is the proportional magnification factor, k d is the differential coefficient, fal(·) is the nonlinear function, u0 is the virtual control variable, u is the actual control variable, b0 is the compensation of the high-order active disturbance rejection control; μ is the adhesion coefficient, θ is the ramp angle, f is the driving resistance, g is the gravity coefficient, and m is the mass of the freight train.

[0022] The present invention establishes a single-particle time-delay model of a freight train, combines the current control amount and the target speed to predict the control amount of the freight train after a certain delay, eliminates the delay in the transmission of the control amount in the freight train system, and achieves the purpose of reducing the overshoot of the speed tracking curve; by increasing the observation of the differential value of the total disturbance in the expansion observer, the accuracy of speed tracking and control is improved, and the accurate estimation of the total disturbance of the freight train when it runs on a complex line is improved; the present invention combines the creep rate observed by the unscented Kalman filter to realize the anti-skid control of the freight train, prevent the freight train from skidding, improve the anti-interference ability of the freight train, and realize the tracking and anti-skid control of the target speed, so as to improve the speed control and operation stability of the freight train.

[0023] Furthermore, the expressions of the velocity estimate, the acceleration estimate, the total disturbance estimate and the differential value of the total disturbance estimate are as follows:

[0024]

[0025] Wherein, e is the output error of the freight train traction braking system, β1, β2, β3, β4 are the adjustable parameters of the high-order extended observer, z1(·) is the speed estimation function, z2(·) is the acceleration estimation function, z3(·) is the total disturbance estimation function, z4(·) is the differential value function of the total disturbance estimation, y(·) is the output speed function of the freight train traction braking system, fal i (·) is a nonlinear function.

[0026] Furthermore, the state quantity X of the unscented Kalman observer is expressed as follows:

[0027] X=[w m w w v z4] T

[0028] The expression of the observation quantity Z of the unscented Kalman observer is as follows:

[0029]

[0030] The expressions of the transformation matrix and observation matrix of the unscented Kalman observer are as follows:

[0031]

[0032] Z(k+1)=B×X(k+1)+v(k)

[0033] Among them, w m is the motor angular velocity, w w is the wheel angular velocity, is the observed value of the longitudinal velocity of the freight train, λ is the creep rate; X(·) is the transformation matrix, Z(·) is the observation matrix, k w is the ratio of the motor angular velocity to the wheel angular velocity, A and B are coefficient matrices, and w m (k) is the motor angular velocity function, w w (k) is the wheel angular velocity function.

[0034] In view of the strong unknown of road surface interference in track contact conditions, which may cause inaccurate observation of creep rate, the present invention introduces the differential value of the total disturbance estimate as input into the unscented Kalman observer, and improves the accuracy of the observed value by predicting the total interference value.

[0035] Furthermore, the expression of the adhesion coefficient μ is as follows:

[0036]

[0037] v=r wv

[0038] sww

[0039] Creep rate λ and slip velocity v s The relationship between them is as follows:

[0040]

[0041] Among them, a, c, d, j are the anti-slip coefficients under different track contact conditions, r w is the wheel radius, v s Represents the slip speed of a freight train, which is the difference between the wheel speed and the longitudinal speed.

[0042] Based on the same inventive concept, the present invention also provides an electronic device, including:

[0043] one or more processors;

[0044] A memory having one or more programs stored thereon, which, when executed by the one or more processors, enables the one or more processors to implement the steps of the freight train speed control method.

[0045] Based on the same inventive concept, the present invention also provides a computer-readable storage medium storing a computer program, which implements the steps of the freight train speed control method when executed by a processor.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention establishes a single-particle time-delay model of a freight train, combines the current control amount and the target speed to predict the control amount of the freight train after a certain delay, eliminates the delay in the transmission of the control amount in the freight train system, and achieves the purpose of reducing the overshoot of the speed tracking curve; by increasing the observation of the differential value of the total disturbance in the expansion observer, the accuracy of speed tracking and control is improved; the present invention combines the creep rate observed by the unscented Kalman filter to realize the anti-skid control of the freight train, prevent the freight train from skidding, improve the anti-interference ability of the freight train, and realize the tracking and anti-skid control of the target speed, thereby improving the speed control and operation stability of the freight train. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is the structural diagram of the high-order active disturbance rejection control;

[0049] Figure 2 This is the control schematic diagram of the ATO system for freight trains;

[0050] Figure 3 It is the creep characteristic curve;

[0051] Figure 4 The figure is a flow chart of the freight train speed control method of the present invention. DETAILED DESCRIPTION

[0052] The present invention will be described in detail below in conjunction with the embodiments. It should be noted that the embodiments and features of the embodiments of the present invention can be combined with each other without conflict. For the convenience of description, if the words "upper", "lower", "left" and "right" appear in the following, they only indicate that the upper, lower, left and right directions are consistent with the drawings themselves, and do not limit the structure.

[0053] Example

[0054] The present embodiment provides a method for controlling the speed of a freight train, which improves the anti-skid control effect of the freight train while achieving the speed tracking of the longitudinal speed of the freight train to the target speed, establishes a single-particle time-delay model of the train, uses a high-order anti-disturbance control method based on input prediction to suppress the influence of the freight train due to large inertia and pure time delay, and uses an untraceable Kalman observer to observe the longitudinal speed and creep rate of the train in real time to prevent wheel slip and idling, thereby improving the anti-interference ability of the freight train and achieving high-precision tracking and anti-skid control of the target speed.

[0055] This embodiment regards the entire train as a rigid individual according to the mechanical characteristics of the freight train itself and the input and output data of actual operation, so that the coupling force between the carriages is no longer explicitly included in the model, and takes the delay existing in the train control process into account, and considers the parallel line and large gentle slope line driving scenarios to establish a single-particle time-delay model of the freight train; based on the single-particle time-delay model, a high-order active disturbance rejection control method is proposed, which generates a transition process and a speed derivative for the input target speed by tracking the differentiator, and the high-order extended state observer estimates each state variable of the speed feedback value, and generates a speed estimate, an acceleration estimate, an estimate of the total disturbance and its Differential value; then the state error feedback device designs the control law for the error value between the tracking differentiator and the output of the extended state observer based on the robust adaptive method, and estimates the real-time action of each state variable and the total disturbance according to the speed value to compensate for the total disturbance, and suppresses the control delay by adding an input prediction module, so as to obtain the actual control amount acting on the freight train; in addition, an unscented Kalman observer is designed to accurately observe the longitudinal speed and creep rate of the freight train, calculate the anti-creep threshold according to the observation content, and compare it with the actual control amount obtained by the high-order active disturbance rejection control to determine the final control amount implemented on the freight train. Accurate speed tracking and anti-skid control of freight trains in the case of idling slip can be achieved.

[0056] like Figure 1-Figure 4 , the freight train speed tracking and anti-creep method includes the following processes:

[0057] Step 1: For the driving scenarios of parallel lines and large gentle slope lines, based on the requirements of fast and stable control, the force conditions of freight trains during driving are analyzed, and a single-particle time-delay model of freight train operation with input time delay is established;

[0058] Step 2: Based on the fact that freight trains basically run on fixed routes, the target speed and route conditions of freight trains are obtained in advance;

[0059] Step 3: Install a hyperspectral camera in front of the freight train to identify the contact conditions of the track and obtain the corresponding Figure 3 The creep characteristic curve is shown.

[0060] Step 4: Based on the real-time operation status information of freight trains, Figure 1 The high-order active disturbance rejection control ADRC shown calculates the actual value of the control variable required for accurate speed tracking of the freight train.

[0061] Step 5: In view of the input delay of the freight train control system due to the large inertia and pure hysteresis, the Figure 1 The input prediction module is used to process the large tracking speed curve overshoot caused by time lag in idling slip conditions.

[0062] Step 6: Use the unscented Kalman observer to observe the longitudinal speed and creep rate of the freight train in real time, and use high-order active disturbance rejection control according to Figure 2 The automatic driving ATO system shown requires it to track the target speed and calculate the anti-creep threshold of the control amount under idling slip conditions.

[0063] Step 7: By comparing the control quantities obtained in step 4 and step 6, and combining the creep characteristic curve, determine the control quantity to be implemented to achieve anti-skid control under the current contact conditions, thereby achieving speed tracking and anti-skid control of the freight train under idling and slipping conditions.

[0064] In the above step 1, the train is regarded as a rigid point to perform mechanical modeling on the train. The single point model of the freight train is expressed as follows:

[0065]

[0066] Among them, x is the real-time position of the freight train, v is the target speed, and F s is the resultant force acting on the freight train, m is the mass of the freight train, and γ is the wheel rotation coefficient of the train.

[0067] The single-particle time-delay model of freight train operation in step 1 above is as follows:

[0068]

[0069] Among them, f(x,v) is the function of the high-order ADRC control system, which is the functional relationship between the real-time position of the freight train and the longitudinal velocity and acceleration. x is the real-time position of the freight train, v is the target velocity, is the acceleration of the freight train, u(k-τ) is the control quantity after delay τ, τ is the delay of the freight train, d(k) is the total disturbance, that is, the sum of the internal disturbance and external disturbance of the train, k is time, b is the compensation of high-order active disturbance rejection control, and y is the output speed of the freight train traction braking system.

[0070] The above step 2 is mainly aimed at the characteristic that freight trains always run back and forth on a fixed line, and the target speed of the line and the specific positions of the ramps, curves and tunnels in the line can be obtained in advance.

[0071] In step 3 above, by using the hyperspectral camera’s ability to identify the contact conditions of the surface of an object when working at a suitable wavelength, the track contact conditions of the freight train running line are identified in real time, and different anti-skid coefficients are selected to obtain a suitable creep characteristic curve, and determine the adhesion threshold of the freight train under the current running conditions. After determining the creep curve, the adhesion coefficient corresponding to the maximum adhesion point is obtained. The adhesion coefficient multiplied by the vertical gravity of the train on the track is the corresponding maximum adhesion (adhesion threshold). The basic creep characteristic curve of the freight train is as follows: Figure 3 As shown, λ represents the creep rate, μ represents the adhesion coefficient, and A represents the maximum adhesion point.

[0072] The adhesion coefficient μ can be expressed by the following formula:

[0073]

[0074] Among them, a, c, d, j are the anti-slip coefficients under different track contact conditions, v s The slip velocity of the freight train, i.e. the difference between the wheel speed and the longitudinal velocity, can be expressed as follows:

[0075] v s =r w w w -v (4)

[0076] Creep rate λ and slip velocity v s The relationship between can be expressed as:

[0077]

[0078] Among them, r w is the wheel radius, w w is the wheel angular velocity.

[0079] The high-order ADRC described in step 4 above is to achieve more accurate tracking. Its specific structure is shown in the figure below. Figure 1 As shown, the main purpose is to increase the acquisition of the differential value of the total disturbance estimate inside and outside the system, as shown below:

[0080]

[0081] Wherein, e is the output error of the freight train traction braking system, z1(·) is the speed estimation function, z2(·) is the acceleration estimation function, z3(·) is the total disturbance estimation function, z4(·) is the differential value function of the total disturbance estimation, y(·) is the output speed function of the freight train traction braking system, b0 is the compensation of the high-order active disturbance rejection control, τ is the delay of the freight train, β1, β2, β3, β4 are the adjustable parameters of the extended observer, fal i (·) is a nonlinear function, and k is time.

[0082] The error feedback control law determines the control quantity according to the nonlinear combination of the system state error. Using the extended state observer, the control quantity is compensated by z3 and z4, and the actual control quantity u can be expressed as:

[0083]

[0084] Among them, v1 is the transition process of the target speed, v2 is the speed derivative, z1 is the speed estimate, z2 is the acceleration estimate, e1 is the first error, e2 is the second error, z3 is the total disturbance estimate, z4 is the differential value of the total disturbance estimate, k p is the proportional magnification factor, k d is the differential term coefficient, fal(·) is the nonlinear function, u0 is the virtual control quantity, and x is the real-time position of the freight train.

[0085] In the above step 5, the control value of the output of the high-order anti-disturbance control is transmitted to the traction brake system of the freight train. Due to the large inertia and pure time lag, the tracking speed curve will be overshot, which is particularly important in the case of idling slip. By adding an input prediction module in the control signal transmission process to suppress the influence of time lag, its specific structure is as follows Figure 1 shown.

[0086] The specific implementation of the unscented Kalman observer in step 6 is as follows:

[0087] In order to strengthen the observation of the longitudinal velocity and creep rate of the freight train in the case of idling slip, the state quantity X in the unscented Kalman observer is defined as:

[0088] X=[w m w w v z4] T (8)

[0089] Among them, w m Indicates the motor angular velocity; w w represents the wheel angular velocity; v represents the target velocity; z4 represents the differential value of the total disturbance estimate.

[0090] In view of the strong uncertainty of road interference in track contact conditions, which may cause inaccurate observation of creep rate, we introduce the differential value of the total disturbance estimate of the high-order extended observer as input into the unscented Kalman observer, and improve the accuracy of the observed value by predicting the total disturbance value.

[0091] In this model, the observation Z is directly defined as the target observation:

[0092]

[0093] in, is the observed real-time longitudinal velocity of the freight train; λ is the creep rate.

[0094] Appropriate selection of state quantities and observation quantities can effectively improve the observation accuracy of the unscented Kalman observer and reduce the difficulty of system observation. The transformation matrix and observation matrix of the model can be defined as:

[0095]

[0096] Z(k+1)=B×X(k+1)+v(k) (11)

[0097] Among them, X(·) is the transformation matrix, Z(·) is the observation matrix, f is the driving resistance, g is the gravity coefficient, θ is the ramp angle, k w is the ratio of the motor angular velocity to the wheel angular velocity, w m is the motor angular velocity, w w is the wheel angular velocity.

[0098] A and B are two corresponding coefficient matrices, which can be expressed as follows:

[0099]

[0100] In step 7, the longitudinal speed and creep rate of the freight train in the case of idling slip can be observed in real time by the untraceable Kalman observer, and the adhesion coefficient μ of the freight train in the case of idling slip can be calculated. At this time, the anti-creep threshold u of the acceleration that the high-order ADRC can output can be calculated based on the maximum adhesion between the wheel and the rail in the case of idling slip. y =μgcosθ-f / m.

[0101] If the actual control amount of the freight train is lower than the anti-creep threshold, it means that the freight train adhesion meets the acceleration requirements, and the train will run according to the actual control amount given by the high-order anti-disturbance control; on the contrary, if the actual control amount of the freight train exceeds the anti-creep threshold, it means that the maximum adhesion of the freight train cannot meet the acceleration requirements at this time, and the freight train will run according to the anti-creep threshold. Therefore, the anti-creep control at this time becomes a speed constraint to prevent slippage.

[0102] The purpose of this embodiment is to improve the speed tracking accuracy and anti-skid control of freight trains when running in idling and slipping conditions. Its features are:

[0103] (1) In order to solve the problems of large inertia and pure time delay of freight trains, a single-particle time-delay model was established based on the driving scenarios of parallel lines and large gentle slope lines to describe the delay of the freight train control system. An input prediction module was used to process it to solve the large tracking speed curve overshoot caused by time delay in idling slip.

[0104] (2) To address the problem that only anti-disturbance control may make inaccurate estimates of system disturbances when freight trains are running in idling and slipping conditions, the speed tracking accuracy is improved by adding observations of the differential value of the total disturbance in the extended observer, thereby improving the accurate estimation of the total disturbance of large-capacity freight train systems running on complex lines.

[0105] (3) Since some anti-skid parameters of freight trains are different under different contact conditions, that is, the corresponding creep characteristic curves are different, in this embodiment, the contact conditions of the track are identified by using a hyperspectral camera, and the applicable creep characteristic curve is determined at this time to provide data support for the subsequent anti-skid control. The hyperspectral camera is used to determine the wheel-rail contact conditions to obtain accurate anti-skid parameters, thereby improving the adaptability of the method.

[0106] (4) An unscented Kalman observer is used to observe the longitudinal speed and creep rate of the freight train in real time. A high-order anti-disturbance control is used to make it track the target speed. The anti-slip threshold of the control variable under idling slip conditions is calculated to prevent the freight train from slipping. This improves the anti-interference ability of the freight train, realizes tracking and anti-slip control of the target speed, and improves the speed control and operation stability of the freight train.

[0107] Another embodiment of the present invention provides an electronic device, including:

[0108] one or more processors;

[0109] A memory having one or more programs stored thereon, which, when executed by one or more processors, enables the one or more processors to implement the steps of the freight train speed control method.

[0110] In some implementations, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0111] In some other implementations, the processor may be a central processing unit (CPU), a digital signal processor (DSP), or other general-purpose processors of various types, which are not limited herein.

[0112] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program, which implements the steps of the freight train speed control method when the computer program is executed by a processor.

[0113] The contents explained in the above embodiments should be understood as these embodiments are only used to more clearly illustrate the present invention, and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

Claims

1. A method for controlling speed of freight trains, characterized in that The implementation process includes the following: Get the target speed of the freight train; Establish a single particle delay model of freight trains, which is shown in the following formula: A tracking differentiator is used to obtain the transition process and velocity derivative of the target velocity; a high-order extended state observer is used to determine a velocity estimate, an acceleration estimate, a total disturbance estimate, and a differential value of the total disturbance estimate; Calculate the error between the transition process and the velocity estimate value as a first error; calculate the error between the velocity derivative and the acceleration estimate value as a second error; A nonlinear error feedback is used to design the control law for the first error and the second error to obtain a virtual control amount; The actual control amount is obtained based on the differential values ​​of the virtual control amount, the total disturbance estimate value and the total disturbance estimate value; the expression of the actual control amount is as follows: A hyperspectral camera is used to identify the track contact conditions of the freight train running line in real time, select the anti-slip coefficient, calculate the creep rate and adhesion coefficient, and obtain the creep characteristic curve; The unscented Kalman observer is used to observe the speed and creep rate of the freight train in real time, and the adhesion coefficient under the current operating conditions of the freight train is obtained according to the creep characteristic curve; the anti-creep threshold is calculated by the adhesion coefficient, and the anti-creep threshold u y The expression is as follows: u y =μgcosθ-f / m If the actual control amount is less than or equal to the anti-creep threshold, the freight train runs according to the actual control amount; if the actual control amount is greater than the anti-creep threshold, the freight train runs according to the anti-creep threshold; Among them, f(x,v) is a function of the higher-order self-immunity control system, x is the real-time position of the freight train, and v is the target speed, is the acceleration of the freight train, u(k-τ) is the control quantity after the delay τ, τ is the delay of the freight train, d(k) is the total disturbance, k is time, b is the compensation of the high-order active disturbance rejection control, y is the output speed of the freight train traction braking system; v1 is the transition process of the target speed, v2 is the speed derivative, z1 is the speed estimate, z2 is the acceleration estimate, e1 is the first error, e2 is the second error, z3 is the total disturbance estimate, z4 is the differential value of the total disturbance estimate, k p is the proportional magnification factor, k d is the differential coefficient, fal(·) is the nonlinear function, u0 is the virtual control variable, u is the actual control variable, b0 is the compensation of the high-order active disturbance rejection control; μ is the adhesion coefficient, θ is the ramp angle, f is the driving resistance, g is the gravity coefficient, and m is the mass of the freight train.

2. The freight train speed control method according to claim 1, characterized in that The expressions of the differential values ​​of the velocity estimate, acceleration estimate, total perturbation estimate and total perturbation estimate are as follows: Wherein, e is the output error of the freight train traction brake system, β1, β2, β3, β4 are the adjustable parameters of the high-order extended state observer, z1(·) is the speed estimation function, z2(·) is the acceleration estimation function, z3(·) is the total disturbance estimation function, z4(·) is the differential value function of the total disturbance estimation, y(·) is the output speed function of the freight train traction brake system, fal i (·) is a nonlinear function.

3. The freight train speed control method according to claim 1, characterized in that The state quantity X expression of the traceless Kalman observer is as follows: X=[in m In w in z4] T The expression of observation Z of the traceless Kalman observer is as follows: The expressions of the transformation matrix and observation matrix of the traceless Kalman Observer are as follows: Z(k+1)=B×X(k+1)+v(k) Among them, w m is the motor angular velocity, w w is the wheel angular velocity, It is the observed value of the longitudinal velocity of the freight train, λ is the creep rate; X (·) is the transformation matrix, Z (·) is the observation matrix, k w It is the ratio of the motor angular velocity to the wheel angular velocity, A and B are coefficient matrix, w m (k) is the motor angular velocity function, w w (k) is the wheel angular velocity function.

4. An electronic device, characterized in that: include: one or more processors; A memory having one or more programs stored thereon, which, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the steps of the method according to any one of claims 1 to 3.

5. A computer-readable storage medium, characterized in that: It stores a computer program, which implements the steps of the method of any one of claims 1-3 when executed by the processor.

Images