CN115128957B - Heavy-duty train operation control method based on iterative learning - Google Patents
Heavy-duty train operation control method based on iterative learning Download PDFInfo
- Publication number
- CN115128957B CN115128957B CN202210857419.3A CN202210857419A CN115128957B CN 115128957 B CN115128957 B CN 115128957B CN 202210857419 A CN202210857419 A CN 202210857419A CN 115128957 B CN115128957 B CN 115128957B
- Authority
- CN
- China
- Prior art keywords
- train
- error
- control
- heavy
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000011159 matrix material Substances 0.000 claims description 31
- 230000006870 function Effects 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 6
- 230000008602 contraction Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000003252 repetitive effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012887 quadratic function Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a heavy-duty train operation control method based on iterative learning, which comprises the following steps of: s1: setting an ideal running speed and an ideal planning position of the train; s2: determining a speed state error, a position state error and a coupler clearance error of the train; s3: obtaining historical level data of the train, and determining a control level according to the historical level data; s4: constructing an iteration control model, and optimizing the control level by using the iteration control model to obtain an optimal control level; s5: and controlling the train by using the optimal control level, updating the actual running speed and the actual running position of the train, judging whether the train reaches the terminal, if so, ending the heavy-duty train running control, and otherwise, returning to the step S2 until the train reaches the terminal. The invention provides an iterative learning control method suitable for heavy-duty train operation control, which is used for fully utilizing historical repeated data so as to improve the heavy-duty train operation control precision.
Description
Technical Field
The invention belongs to the technical field of heavy-load train control, and particularly relates to a heavy-load train operation control method based on iterative learning.
Background
Heavy-duty trains have become a popular transportation mode due to their large capacity, low cost, and high efficiency. With the gradual expansion of the running scale of the heavy-duty train and the increase of the vehicle-mounted weight, the increase of the number of the carriages and the increase of the length of the train, the caused excessive longitudinal impact force is a main factor causing great difficulty in the operation of the heavy-duty train. Therefore, the heavy-duty train is difficult to realize rapid and accurate tracking control, and even accidents such as hook breaking and derailment are caused by large coupler force between trains. Therefore, it is necessary to construct a more accurate longitudinal dynamics model and design a controller with higher accuracy to improve the operation control performance of the heavy-duty train.
Trains are typically operated periodically according to a given speed profile according to a schedule established by the railroad department. The heavy-duty train has the characteristics of repeatability of line conditions, periodicity of wheel rotation, similarity of running environments of a plurality of trains in a short time, invariance of a train kinematic model, consistency of vehicle characteristics and the like in the special freight line running process. The driver manually operates the train according to personal experience, which may result in a low control accuracy. Aiming at the problems, the repeatability of the system is effectively utilized by using iterative learning control for realizing rapid and high-precision tracking, and the model parameters and model uncertainty of the system are solved. The iterative learning control which is essentially model-free is suitable for a nonlinear control system like train control. The method is not applied to the heavy-load train with highly coupled and complex interior. Therefore, an iterative learning control method suitable for heavy-duty train operation control is provided.
A train of two ten thousand tons of heavy-duty trains consists of a plurality of locomotives and two hundred or more trailers, and under the condition of considering the longitudinal impulse between carriages, the limited control quantity is difficult to simultaneously ensure that the serial coupling force between multiple carriages of an overlong and overweight marshalling meets the safety requirement. The too high order and the nonlinearity of the model make it difficult to combine the feedback control mode to control the multi-particle operation of the heavy-duty train. Therefore, the iterative learning control method based on the multi-quality-point model has theoretical research value for the operation control of the heavy-duty train. In the heavy-duty train operation control process, accurate modeling is difficult, various uncertain complex disturbances exist, the disturbances include repetitive disturbances caused by high repetition of train operation characteristics, and non-repetitive disturbances caused by inconsistent operation conditions. Iterative learning can well utilize repeatability information in the system and realize whole-process tracking control, and model prediction control can timely and dynamically respond to non-repeatability disturbance. The model predictive control has efficient dynamic control performance, including predictive model, rolling optimization and feedback correction. The model prediction control can timely make up uncertainty caused by factors such as model mismatch, distortion and interference. Therefore, the heavy-duty train feedback control method based on the model prediction control method has an application prospect.
In conclusion, the tracking control of the heavy-duty train is a comprehensive and complex problem, wherein the model prediction control and the iterative learning control have own advantages and disadvantages. Therefore, the invention provides the improved iterative learning model predictive control heavy-load train operation control method with better tracking performance by combining the model predictive control and the iterative learning.
Disclosure of Invention
The invention aims to solve the problem that a multi-quality-point model in the existing feedback control method is difficult to apply to a heavy-duty train, and provides a heavy-duty train operation control method based on iterative learning.
The technical scheme of the invention is as follows: a heavy-duty train operation control method based on iterative learning comprises the following steps:
s1: setting an ideal running speed and an ideal planning position of the train;
s2: acquiring the actual running speed and the actual running position of the train, and determining the speed state error, the position state error and the coupler clearance error of the train according to the ideal running speed and the ideal planning position of the train and the actual running speed and the actual running position of the train;
s3: obtaining historical level data of the train, and determining a control level according to the historical level data;
s4: constructing an iteration control model, and optimizing the control level by using the iteration control model to obtain an optimal control level;
s5: and controlling the train by using the optimal control level, updating the actual running speed and the actual running position of the train, judging whether the train reaches the terminal, if so, ending the heavy-duty train running control, and otherwise, returning to the step S2 until the train reaches the terminal.
Further, in step S2, the speed state error of the vehicle i
Error in position state
Clearance error of car coupler
The calculation formulas of (A) and (B) are respectively as follows:
wherein v is r (t) is the target speed of the head car at time t, v i (t) is the actual speed of the car i at time t, s r (t) is the target position of the first carriage at the time t; s i (t) is the actual position of the car i at time t, s i+1 (t) is the actual position of the car i +1 at time t; l i When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i and the first carriage is l i+1 When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i +1 and the first carriage,
is the position state error of the vehicle compartment i + 1.
Further, step S4 comprises the following sub-steps:
s41: constructing an error state space expression;
s42: discretizing an error state space expression and constructing a target function of a fixed time domain;
s43: calculating a minimum optimal solution of an objective function of a fixed time domain;
s44: constructing an iterative control model according to the minimum optimal solution of the objective function of the fixed time domain;
s45: and optimizing the control level by using the iteration control model to obtain the optimal control level.
Further, in step S41, the error state space expression is:
Y(t)＝Ce(t)
wherein,
denotes the derivative of speed with position, Y (t) denotes the output state, a denotes the first coefficient matrix, B denotes the second coefficient matrix, C denotes the third coefficient matrix, e (t) denotes the error state vector, Δ u (t) denotes the error control vector, and w' (t) denotes the process noise.
Further, in step S42, the expression of the objective function J in the fixed time domain is:
wherein T represents the predicted time, N represents the number of vehicles, k v Representing the velocity weight, k s Representing the position weight, k x The weight of the inter-vehicle distance is represented,
the speed difference of the vehicle compartment i is indicated,
indicating the error in the positional state of the vehicle compartment i,
the error indicates the error in the position state of the vehicle i +1, Δ u (j) indicates the change in force at time j, and R indicates the control force change coefficient.
Further, in step S43, the equation for calculating the minimum optimal solution Δ u of the objective function in the fixed time domain is:
Δu＝-(R+B T PB) -1 B T PAe(k)＝-Ke(k)
wherein A represents a first coefficient matrix, B represents a second coefficient matrix, K represents a state feedback gain matrix, R represents a control force variation coefficient, P represents a symmetric positive definite constant matrix, and e (K) represents an error state vector.
Further, in step S44, the iterative control model u k The expression of (t) is:
wherein,
a model-predictive control increment is represented,
represents the iterative learning rate, A represents the first coefficient matrix, B represents the second coefficient matrix, k m Representing model predictive control gain weights, R representing control force variation coefficients, P representing a symmetric positive definite constant matrix, e k (t) represents the error state vector at time t in the kth iteration,
represents the last iterative learning rate, alpha represents the current iterative error and the last iterative error weight, gamma represents the learning gain,
representing the error state vector at time t in the (k-1) th iteration,
representing the derivative of the error state vector at time t in the kth iteration.
The invention has the beneficial effects that:
(1) The invention provides an iterative learning control method suitable for heavy-duty train operation control, which is used for fully utilizing historical repeated data so as to improve the operation control precision of a heavy-duty train;
(2) The heavy-duty train operation control method based on the multi-mass-point model is beneficial to reducing the longitudinal impulse of the train and ensuring the safe and stable operation of the heavy-duty train;
(3) By the controller integrating model predictive control and iterative learning design, the method can also help to inhibit the influence of non-repetitive disturbance and improve the accuracy and reliability of control.
Drawings
FIG. 1 is a flow chart of a method for controlling the operation of a heavy haul train;
FIG. 2 is a schematic diagram of an application of the heavy haul train operation control method;
fig. 3 is a diagram of a controller architecture based on improved iterative learning model predictive control.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a heavy haul train operation control method based on iterative learning, which comprises the following steps:
s1: setting an ideal running speed and an ideal planning position of the train;
s2: acquiring the actual running speed and the actual running position of the train, and determining the speed state error, the position state error and the coupler clearance error of the train according to the ideal running speed and the ideal planning position of the train and the actual running speed and the actual running position of the train;
s3: acquiring historical level data of the train, and determining a control level according to the historical level data;
the control quantity exists in the form of continuous force and needs to be converted into a discrete level control mode to operate the heavy-duty train. In order to avoid reducing the passenger's riding experience due to frequent level switching, it is necessary to output an appropriate control level according to the historical level data.
S4: constructing an iteration control model, and optimizing the control level by using the iteration control model to obtain an optimal control level;
s5: and controlling the train by using the optimal control level, updating the actual running speed and the actual running position of the train, judging whether the train reaches the terminal, if so, ending the heavy-duty train running control, and otherwise, returning to the step S2 until the train reaches the terminal.
As shown in FIG. 2 and FIG. 3, the controller is used for heavy-duty train trajectory tracking and is based on improved iterative learning model predictive control, the input of the controller is the tracking state error of the train, and the controlled variable is obtained after the output of the model predictive controller and the iterative learning law is weighted.
And (3) state error: the train tracks the difference between the desired speed, position and the actual speed, position of the train.
Train operating speed, position tracking expectation: and calculating an ideal speed and position curve according to parameters such as train characteristics, line characteristics and the like of the ideal train model.
An ideal train model: and obtaining an ideal train model by train design parameters.
Tracking a speed curve: because the actual train has a certain deviation from the ideal train model, the deviation occurs when the train runs according to the train level planned by the ideal running curve, so the control is carried out to adjust the actual running speed curve of the train to be close to the planned speed curve of the train as much as possible, namely the train speed curve tracking.
Level: the handle level is used for controlling the train to drag or brake.
In the embodiment of the present invention, in step S2, the objective of the operation control is to find an appropriate control amount to make the output state approach to the ideal output and eliminate the consistency error. The speed difference of adjacent cars and the position difference of the buffer are main influence indexes influencing coupler force. In order to reduce the risk of unhooking and unhooking of a heavy-duty train caused by excessive coupler force, the speeds of adjacent carriages need to be kept consistent and the telescopic state of a buffer needs to be kept to be zero as much as possible. Speed state error of car i
Error in position state
Clearance error of car coupler
The calculation formulas of (A) and (B) are respectively as follows:
wherein v is r (t) is the target speed of the head car at time t, v i (t) is the actual speed of the car i at time t, s r (t) is the target position of the first carriage at the time t; s i (t) is the actual position of the car i at time t, s i+1 (t) is the actual position of the car i +1 at time t; l i When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i and the first carriage is l i+1 When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i +1 and the first carriage,
is the position state error of the vehicle compartment i + 1.
In an embodiment of the present invention, step S4 includes the following sub-steps:
s41: constructing an error state space expression;
s42: discretizing an error state space expression and constructing a target function of a fixed time domain;
s43: calculating a minimum optimal solution of the target function of the fixed time domain;
s44: constructing an iterative control model according to the minimum optimal solution of the objective function of the fixed time domain;
s45: and optimizing the control level by using the iteration control model to obtain the optimal control level.
In the embodiment of the invention, in step S41, in order to solve the problem that the multi-mass-point model in the existing feedback control method is difficult to be applied to heavy-duty trains, the integral of the speed states and the position states of all vehicles is used as a performance index, and the performance index is converted into a linear quadratic optimal control problem based on a model predictive control framework. In order to construct a quadratic performance index conveniently, an error state space expression is constructed as follows:
Y(t)＝Ce(t)
wherein,
denotes the derivative of speed with position, Y (t) denotes the output state, a denotes the first coefficient matrix, B denotes the second coefficient matrix, C denotes the third coefficient matrix, e (t) denotes the error state vector, Δ u (t) denotes the error control vector, and w' (t) denotes the process noise.
In the embodiment of the invention, in step S42, in the heavy-duty train operation control, the accuracy of train tracking is improved and the coupler force is reduced by tracking the target speed and the target position of each vehicle. Discretizing the state equation of the continuous time domain, and constructing an expression of an objective function J of the fixed time domain as follows:
wherein T represents the predicted time, N represents the number of vehicles, k v Representing the velocity weight, k s Representing the position weight, k x The weight of the inter-vehicle distance is represented,
the speed difference of the vehicle compartment i is indicated,
indicating the error in the positional state of the vehicle compartment i,
the error indicates the error in the position state of the vehicle i +1, Δ u (j) indicates the change in force at time j, and R indicates the control force change coefficient.
In the embodiment of the present invention, in step S43, a calculation formula of the minimum optimal solution Δ u of the objective function in the fixed time domain is as follows:
Δu＝-(R+B T PB) -1 B T PAe(k)＝-Ke(k)
wherein A represents a first coefficient matrix, B represents a second coefficient matrix, K represents a state feedback gain matrix, R represents a control force variation coefficient, P represents a symmetric positive definite constant matrix, and e (K) represents an error state vector.
In the embodiment of the present invention, in step S44, the iterative learning model predictive control law is determined by the iterative learning rate
And model predictive control increments
The iterative learning and the model predictive control are mutually independent, so that the feedback control quantity generated by the model predictive control is not used as the experience information of the next iteration of the iterative learning, and the model u is controlled by the iterative control k The expression of (t) is:
Γ(t)C(t)B(t)＝k l (R+B T PB) -1 B T PACB
wherein,
the model is represented as a predictive control increment,
represents the iterative learning rate, A represents the first coefficient matrix, B represents the second coefficient matrix, k m Representing model predictive control gain weight, R representing control force variation coefficient, P representing symmetric positive definite constant matrix, e k (t) represents the error state vector at time t in the kth iteration,
represents the last iterative learning rate, alpha represents the current iterative error and the last iterative error weight, gamma represents the learning gain,
representing the error state vector at time t in the (k-1) th iteration,
representing the derivative of the error state vector at time t in the kth iteration.
The model prediction control and the quadratic optimal control both adopt a performance index control method taking a quadratic function as integral. The quadratic optimal solution is an essential condition for gradually stabilizing the closed-loop system, so that the model predicts the control increment
The design is as follows:
the open-close loop iterative learning control algorithm is designed according to the error of the current iteration and the error of the last iteration. And adoptIs the last iterative learning rate
Rather than the last control force u k-1 (t), which ensures independence of iterative learning from model predictive control. And an iterative learning control method suitable for the heavy-duty train is designed based on quadratic optimal control.
The working principle and the process of the invention are as follows: iterative learning can well utilize the system operation repeatability characteristics of the heavy-duty train and realize whole-course tracking control, and model prediction control can timely and dynamically respond to non-repeatability disturbance, so the invention provides an improved iterative learning model prediction control method suitable for the heavy-duty train, thereby improving the accurate tracking control of the heavy-duty train and relieving longitudinal impulse.
The beneficial effects of the invention are as follows:
(1) The invention provides an iterative learning control method suitable for heavy-duty train operation control, which is used for fully utilizing historical repeated data so as to improve the heavy-duty train operation control precision;
(2) The heavy-duty train operation control method based on the multi-mass-point model is beneficial to reducing the longitudinal impulse of the train and ensuring the safe and stable operation of the heavy-duty train;
(3) By the controller integrating model predictive control and iterative learning design, the method can also contribute to inhibiting the influence of non-repetitive disturbance and improving the accuracy and reliability of control.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (6)
1. A heavy-duty train operation control method based on iterative learning is characterized by comprising the following steps:
s1: setting an ideal running speed and an ideal planning position of the train;
s2: acquiring the actual running speed and the actual running position of the train, and determining the speed state error, the position state error and the coupler clearance error of the train according to the ideal running speed and the ideal planning position of the train and the actual running speed and the actual running position of the train;
s3: obtaining historical level data of the train, and determining a control level according to the historical level data;
s4: constructing an iteration control model, and optimizing the control level by using the iteration control model to obtain an optimal control level;
s5: controlling the train by using the optimal control level, updating the actual running speed and the actual running position of the train, judging whether the train reaches the end point, if so, ending the heavy-duty train running control, otherwise, returning to the step S2 until the train reaches the end point;
in the step S2, the speed state error of the carriage i
Error in position state
Clearance error of car coupler
The calculation formulas of (A) and (B) are respectively as follows:
wherein v is r (t) is the target speed of the head car at time t, v i (t) is the actual speed of the car i at time t, s r (t) is the target position of the first carriage at the time t; s i (t) is the actual position of the car i at time t, s i+1 (t) is the actual position of the car i +1 at time t; l. the i When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i and the first carriage is l i+1 When the expansion and contraction state of the buffer of all the trains is zero, the distance between the carriage i +1 and the first carriage,
is the position state error of the vehicle compartment i + 1.
2. The heavy-duty train operation control method based on iterative learning of claim 1, wherein said step S4 comprises the substeps of:
s41: constructing an error state space expression;
s42: discretizing an error state space expression and constructing a target function of a fixed time domain;
s43: calculating a minimum optimal solution of the target function of the fixed time domain;
s44: constructing an iterative control model according to the minimum optimal solution of the objective function of the fixed time domain;
s45: and optimizing the control level by using the iteration control model to obtain the optimal control level.
3. The heavy-duty train operation control method based on iterative learning of claim 2, wherein in said step S41, the error state space expression is:
Y(t)＝Ce(t)
wherein,
denotes the derivative of speed with position, Y (t) denotes the output state, a denotes the first coefficient matrix, B denotes the second coefficient matrix, C denotes the third coefficient matrix, e (t) denotes the error state vector, Δ u (t) denotes the error control vector, and w' (t) denotes the process noise.
4. The heavy-duty train operation control method based on iterative learning of claim 2, wherein in said step S42, an expression of an objective function J of a fixed time domain is:
wherein T represents the predicted time, N represents the number of vehicles, k v Representing the velocity weight, k s Representing the position weight, k x The weight of the inter-vehicle distance is represented,
the speed difference of the vehicle compartment i is indicated,
indicating the error in the positional state of the vehicle compartment i,
the error indicates the error in the position state of the vehicle i +1, Δ u (j) indicates the change in force at time j, and R indicates the control force change coefficient.
5. The method for controlling the operation of the heavy haul train based on the iterative learning of claim 2, wherein in the step S43, the formula for calculating the minimum optimal solution Δ u of the objective function in the fixed time domain is as follows:
Δu＝-(R+B T PB) -1 B T PAe(k)＝-Ke(k)
wherein A represents a first coefficient matrix, B represents a second coefficient matrix, K represents a state feedback gain matrix, R represents a control force variation coefficient, P represents a symmetric positive definite constant matrix, and e (K) represents an error state vector.
6. The heavy-duty train operation control method based on iterative learning of claim 2, wherein in said step S44, the iterative control model u is used k The expression of (t) is:
wherein,
the model is represented as a predictive control increment,
represents the iterative learning rate, A represents the first coefficient matrix, B represents the second coefficient matrix, k m Representing model predictive control gain weights, R representing control force variation coefficients, P representing a symmetric positive definite constant matrix, e k (t) represents the error state vector at time t in the kth iteration,
represents the last iterative learning rate, alpha represents the current iterative error and the last iterative error weight, gamma represents the learning gain,
representing the error state vector at time t in the (k-1) th iteration,
the derivative of the error state vector at time t in the kth iteration is represented and K represents the state feedback gain matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210857419.3A CN115128957B (en) | 2022-07-20 | 2022-07-20 | Heavy-duty train operation control method based on iterative learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210857419.3A CN115128957B (en) | 2022-07-20 | 2022-07-20 | Heavy-duty train operation control method based on iterative learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115128957A CN115128957A (en) | 2022-09-30 |
| CN115128957B true CN115128957B (en) | 2023-03-14 |
Family
ID=83384557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210857419.3A Active CN115128957B (en) | 2022-07-20 | 2022-07-20 | Heavy-duty train operation control method based on iterative learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115128957B (en) |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105446320B (en) * | 2015-12-17 | 2018-02-16 | 南京航空航天大学 | A kind of fault detection method of the vertical suspension of bullet train based on limited frequency domain |
| US10093331B2 (en) * | 2016-08-08 | 2018-10-09 | Mitsubishi Electric Research Laboratories, Inc. | Train automatic stopping control with quantized throttle and braking |
| CN107640183B (en) * | 2017-07-31 | 2020-08-25 | 北京石油化工学院 | Train operation control method based on iterative learning control |
| CN110083056A (en) * | 2019-05-28 | 2019-08-02 | 北京交通大学 | A kind of municipal rail train PID controller automatic adjusting method |
| US11640528B2 (en) * | 2019-10-22 | 2023-05-02 | Baidu Usa Llc | Method, electronic device and computer readable medium for information processing for accelerating neural network training |
| CN110824465B (en) * | 2019-10-25 | 2022-01-04 | 中铁武汉勘察设计研究院有限公司 | Method and system for positioning molten iron combined transport vehicle based on image recognition and radar measurement |
| CN112193280B (en) * | 2020-12-04 | 2021-03-16 | 华东交通大学 | Heavy-load train reinforcement learning control method and system |
| CN112590738B (en) * | 2020-12-23 | 2022-03-08 | 交控科技股份有限公司 | ATO (automatic train operation) parking control method compatible with different inter-vehicle generations |
| CN113085963B (en) * | 2021-04-22 | 2021-09-14 | 西南交通大学 | Dynamic regulation and control method and dynamic regulation and control device for train control level |
| CN113552801B (en) * | 2021-07-08 | 2024-04-12 | 北京交通大学 | Virtual formation operation control method based on distributed subway train |
| CN113911172A (en) * | 2021-10-12 | 2022-01-11 | 中车大连机车研究所有限公司 | High-speed train optimal operation control method based on self-adaptive dynamic planning |
| CN114326386B (en) * | 2021-11-30 | 2024-01-23 | 卡斯柯信号有限公司 | Automatic train driving track planning and tracking integrated control method and device |
-
2022
- 2022-07-20 CN CN202210857419.3A patent/CN115128957B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115128957A (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20040238693A1 (en) | Control system for operating long vehicles | |
| CN104590333A (en) | Railway train intelligent operation optimization control system | |
| CN110281982B (en) | Heavy-load train cruise control method based on periodic intermittent control | |
| Graffione et al. | Model predictive control of a vehicle platoon | |
| CN109703606B (en) | High-speed train intelligent driving control method based on historical operation data | |
| CN106056238B (en) | Planning method for train interval running track | |
| CN112782978B (en) | High-speed train cruising operation control method based on self-triggering mechanism | |
| CN116080725B (en) | Virtual reconnection high-speed train tracking interval control method based on model prediction | |
| CN112507464A (en) | Freight train operation curve optimization method based on improved multi-target wolf algorithm | |
| CN111580391B (en) | Motor train unit traction torque control method based on model prediction | |
| CN114942642A (en) | Unmanned automobile track planning method | |
| CN116594305A (en) | Virtual marshalling train tracking control system and method based on model predictive control | |
| Liu et al. | An approach for accurate stopping of high-speed train by using model predictive control | |
| CN114326386A (en) | Automatic train driving track planning and tracking integrated control method and device | |
| Zhang et al. | Braking-penalized receding horizon control of heavy-haul trains | |
| CN115128957B (en) | Heavy-duty train operation control method based on iterative learning | |
| Lang et al. | Dqn-based speed curve optimization for virtual coupling | |
| Yang et al. | Optimal Operation of High‐Speed Trains Using Hybrid Model Predictive Control | |
| CN108749816B (en) | Method for regulating and controlling speed of intelligent vehicle by using energy dissipation theory | |
| CN112109773B (en) | Train speed control method based on H infinity and multi-objective optimization | |
| CN117170228A (en) | Self-adaptive sliding mode control method for virtual marshalling high-speed train interval control | |
| CN116339190B (en) | Train operation control method based on non-reference track model predictive control | |
| CN114274787A (en) | Magnetic suspension train operation control method based on adaptive neural network | |
| CN114625003A (en) | High-speed train distributed track tracking control method based on multi-quality-point model | |
| CN115524975A (en) | Virtual coupling strategy and model prediction control method for train operation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |