Virtual marshalling train minimum tracking spacing generation method based on parameter identification

Abstract

The invention provides a virtual marshalling train minimum tracking spacing generation method based on parameter identification. The method comprises the following steps: acquiring actual operation data of a virtual marshalling train, wherein the actual operation data comprises train operation time, train operation speed, train mass and train braking force; establishing a virtual marshalling train dynamic model by considering the braking force, the basic resistance and the random interference of the train, and performing parameter identification on the virtual marshalling train dynamic model by applying a polynomial regression algorithm based on the actual operation data of the train; and predicting a braking track of the virtual marshalling train based on the dynamic model of the virtual marshalling train after parameter identification, and obtaining the minimum tracking distance of the virtual marshalling train based on the braking tracks of the front train and the rear train in the virtual marshalling train. According to the method, the improvement of a virtual marshalling train control strategy is realized, and compared with a method for realizing a relative braking distance by adopting a fixed braking rate, the method has the advantages that the train tracking distance is shortened, and more efficient and safer train operation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail transit, and in particular, to a method for generating the minimum tracking interval of a virtual formation train based on parameter identification. Background Art

[0002] With the rapid development of China's economy and the continuous acceleration of the urbanization process, the contradiction between the growing urban traffic demand and the limited urban traffic supply has become increasingly prominent, and the problem of traffic congestion has become an urgent problem to be solved in these cities. As the backbone of urban public transportation, rail transit lines have become the best way to solve traffic congestion and ensure the sustainable development of cities.

[0003] With the expansion of the urban rail transit network scale, the characteristics of the unbalanced spatio-temporal distribution of urban rail transit passenger flow have become more and more obvious, bringing challenges such as tight transport capacity in some line sections during peak periods and low utilization rates of a large number of lines and vehicles during off-peak periods to the operation of urban rail transit. Train virtual formation relying on the full-automatic operation technology is considered an effective means to solve or alleviate the above problems. Train virtual formation can achieve precise and flexible transport capacity adjustment through the efficient and flexible utilization of vehicle resources, solve the problem of the unbalanced spatio-temporal distribution of urban rail transit passenger flow, and thus improve the operation service quality and reduce the operation cost.

[0004] Virtual formation does not rely on mechanical connection devices, but through technologies such as wireless communication and automatic control, multiple trains (two trains) are "coupled" with a small enough interval to run in a formation, greatly shortening the tracking interval between the front and rear trains, and this coupling can be dynamically "uncoupled" online as needed, thus greatly improving the efficiency and flexibility of train formation adjustment. At the same time, it poses higher requirements for the train control under virtual formation.

[0005] The parameter identification of urban rail trains under traditional fixed formation accurately identifies the basic resistance parameters, traction and braking parameters, etc. of the trains, designs a controller with better performance, so that the trains can run more precisely according to the predetermined speed and position during operation, reducing errors and delays.

[0006] At present, there is no effective method for generating the tracking interval of urban rail trains under virtual formation in the prior art. Summary of the Invention

[0007] The present invention provides a method for generating the minimum tracking interval of a virtual formation train based on parameter identification to effectively shorten the train tracking interval and achieve more efficient operation of the virtual formation train.

[0008] To achieve the above object, the present invention adopts the following technical solutions.

[0009] A method for generating the minimum tracking interval of a virtual formation train based on parameter identification, comprising:

[0010] Obtain the actual operation data of the virtual formation train, where the actual operation data includes the train operation time, train operation speed, train mass, and train braking force;

[0011] Construct a virtual formation train dynamics model considering the train braking force, basic resistance, and random interference, and perform parameter identification on the virtual formation train dynamics model using the polynomial regression algorithm based on the actual operation data of the train;

[0012] Predict the braking trajectory of the virtual formation train based on the virtual formation train dynamics model after parameter identification, and obtain the minimum tracking interval of the virtual formation train based on the braking trajectories of the front and rear trains in the virtual formation train.

[0013] Preferably, the obtaining of the actual operation data of the virtual formation train, where the actual operation data includes the train operation time, train operation speed, train mass, and train braking force, includes:

[0014] Acquire the actual operation data during the braking of the virtual formation train by collecting the operation process of the virtual formation train on-site, where the actual operation data includes the train operation time, train operation speed, train mass, and train braking force.

[0015] Preferably, the constructing of the virtual formation train dynamics model considering the train braking force, basic resistance, and random interference, and performing parameter identification on the virtual formation train dynamics model using the polynomial regression algorithm based on the actual operation data of the train, includes:

[0016] Establish a virtual formation train dynamics model considering the braking force, basic resistance, and random interference during the train braking process as follows:

[0017]

[0018] In the formula, x i (t) represents the displacement of each train in the virtual formation; v i (t) represents the speed of each train in the virtual formation;

[0019] u i (t) represents the control acceleration of the train; F U (t) represents the traction force F QY (t) or the braking force F ZD (t); δ i (t) represents the external force per unit mass of each train, including the basic resistance F JB (t) and other additional resistances F FJ(t); m represents the mass of the train; g is the acceleration due to gravity; a, b, and c are drag coefficients;

[0020] Discretize the dynamic model of the virtual formation train into:

[0021]

[0022] Set an nth-order polynomial regression model as:

[0023] y = β0 + β1x + β2x 2 +... + β n x n + ε(4)

[0024] y is the target variable, x is the independent variable, β0 and β1 are model parameters, and ε is the error term;

[0025] Apply the actual running data of the train, time t, speed v(t), train mass m, and train braking force F ZD (t)), and the dynamic model of the virtual formation train shown in Equation (4) to the polynomial regression algorithm shown in Equation (4) for parameter identification;

[0026] Then the polynomial regression model is changed to:

[0027] Among them,

[0028] Use this polynomial regression model to identify the drag parameters a, b, and c in the dynamic model of the virtual formation train.

[0029] Construct the x 2 feature in the polynomial regression model, and form a design matrix with the original feature x, expressed as: Then the polynomial regression model can be expressed as:

[0030] Complete parameter identification for this regression model through linear regression to obtain the identified parameters a, b, and c;

[0031] Finally, calculate y based on the obtained identified parameters a, b, and c, which is the identified acceleration a i '(t) of the train, then the identified speed at any time of the train is v i '(t) = v i (0) + ta i '(t).

[0032] Preferably, predicting the braking trajectory of the virtual formation train based on the dynamic model of the virtual formation train after parameter identification, and obtaining the minimum tracking interval of the virtual formation train based on the braking trajectories of the front and rear trains in the virtual formation train, including:

[0033] The train identification speed v is obtained from the train dynamics model based on the identification parameters i '(t). According to the train identification speed v i '(t) and the train running time t, the train braking trajectory is predicted. Based on the predicted braking trajectory of the leading train, the speed v2'(t) during the braking of the leading train and the position of the rear end of the leading train at any time are obtained s2 (t);

[0034] Set the safety constraint conditions as:

[0035]

[0036] In the formula, s 2 represents the position of the rear end of the leading train at the parking moment; represents the position of the front end of the trailing train at the parking moment; s represents the protection distance; s2 (t) represents the position of the rear end of the leading train at any time; represents the position of the front end of the trailing train at any time; v1'(t) represents the speed during the braking of the trailing train; a(t) represents the train acceleration;

[0037] According to the identification speed v2'(t) of the leading train and combined with the safety constraint conditions, the speed v1'(t) of the trailing train at any time is predicted, and the predicted braking trajectory of the trailing train is generated. According to the predicted braking trajectories of the front and rear trains in the virtual formation, the initial interval distance between the front and rear trains in the virtual formation is calculated, and this initial interval distance is used as the minimum tracking interval of the virtual formation train.

[0038] As can be seen from the technical solutions provided by the embodiments of the present invention above, the method for generating the minimum tracking interval of virtual formation trains based on parameter identification proposed by the present invention improves the train operation efficiency and is of great significance to the sustainable development of urban rail transit systems. First, this method collects the actual operation data of trains, constructs a train dynamics model, and uses the polynomial regression algorithm to achieve high-precision prediction of the train braking trajectory, improving the train control accuracy and safety. During the optimization process, the present invention takes the predicted train braking trajectory as the braking trajectory of the leading train, and dynamically adjusts and generates the braking trajectory of the trailing train under the consideration of safety constraint conditions, shortening the tracking interval between train units and improving the line passing capacity and train operation efficiency. Through iterative optimization, the present invention has successfully realized the improvement of the virtual formation train control strategy. Compared with the method of using a fixed braking rate to achieve the relative braking distance, the train tracking interval is shortened, and a more efficient and safe train operation is achieved.

[0039] The additional aspects and advantages of the present invention will be given in part in the following description, and these will become obvious from the following description or be understood through the practice of the present invention. Brief Description of the Drawings

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0041] Figure 1 It is a processing flowchart of a method for generating the minimum tracking interval of a virtual formation train based on parameter identification provided by an embodiment of the present invention;

[0042] Figure 2 It is an actual train operation data diagram during braking provided by an embodiment of the present invention;

[0043] Figure 3 It is a comparison diagram of the actual braking trajectory and the identified braking trajectory provided by an embodiment of the present invention;

[0044] Figure 4 It is a processing flowchart of generating the braking trajectory of the rear vehicle considering safety constraints based on the braking trajectory of the front vehicle provided by an embodiment of the present invention;

[0045] Figure 5 It is a comparison diagram of the optimized braking trajectories of the front and rear vehicles in the virtual formation provided by an embodiment of the present invention. Specific Embodiments

[0046] The following details the embodiments of the present invention. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be construed as a limitation of the present invention.

[0047] Those skilled in the art of this technology can understand that unless specifically stated, the singular forms "a", "an", "the" and "said" used here may also include the plural forms. It should be further understood that the term "comprising" used in the description of the present invention means the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used here may include wireless connection or coupling. The phrase "and / or" used here includes any unit and all combinations of one or more related listed items.

[0048] Those skilled in the art can understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those of ordinary skill in the art to which this invention belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined as herein.

[0049] For the convenience of understanding the embodiments of the present invention, the following will further explain with several specific embodiments in conjunction with the accompanying drawings, and each embodiment does not constitute a limitation to the embodiments of the present invention.

[0050] The processing flow chart of a method for generating the minimum tracking interval of a virtual formation train based on parameter identification provided by an embodiment of the present invention is as Figure 1 shown, including the following processing steps;

[0051] Step S10: Obtain the actual operation data of the virtual formation train, where the actual operation data includes train operation time, train operation speed, train mass, train braking force, etc.

[0052] Step S20: Consider the train braking force, basic resistance and random interference to construct a virtual formation train dynamics model, and use the polynomial regression algorithm to identify the parameters of the virtual formation train dynamics model based on the actual operation data of the train.

[0053] Step S30: Predict the braking trajectory of the virtual formation train based on the parameter identification result.

[0054] Step S40: Obtain the minimum tracking interval of the virtual formation train based on the braking trajectories of the front and rear trains in the virtual formation train.

[0055] For the purpose of minimizing the tracking interval as much as possible, use the predicted train braking trajectory as the front train braking trajectory, and consider safety constraints to dynamically adjust and generate the rear train braking trajectory.

[0056] If the target parking position is reached and the safety constraints are met, end; otherwise, return to "dynamically adjust the rear train braking trajectory according to the front train braking trajectory" to continue iterative optimization.

[0057] Finally, output the predicted rear train braking trajectory. Based on the braking trajectories of the front and rear trains, the minimum tracking interval of the virtual formation train based on parameter identification can be obtained. Compared with the traditional method based on relative braking distance, the train tracking interval is smaller and the train operation efficiency is higher.

[0058] The above step S10 includes: initialization and data input, obtain the actual operation data during train braking through on-site collection, and draw the actual braking trajectory s-t diagram of the train. Figure 2An actual operation data diagram during train braking provided by an embodiment of the present invention.

[0059] The above step S20 includes: considering factors such as braking force, basic resistance, and random interference during the train braking process, and establishing a virtual formation train dynamics model as follows:

[0060]

[0061] In the formula, x i (t) represents the displacement of each train in the virtual formation; v i (t) represents the speed of each train in the virtual formation; u i (t) represents the control acceleration of the train; F U (t) represents the traction force F QY (t) or the braking force F ZD (t); δ i (t) represents the external force received by each train per unit mass, including the basic resistance F JB (t) and other additional resistances F FJ (t) related to the environment; m represents the train mass; g is the acceleration due to gravity; a, b, c are resistance coefficients, generally taken according to experience..

[0062] Discretize the above virtual formation train dynamics model as:

[0063]

[0064] Construct a polynomial regression parameter identification algorithm. For a univariate polynomial regression, if the original linear model is y = β0 + β1x + ε, where y is the target variable, x is the independent variable, β0 and β1 are the model parameters, and ε is the error term, then a quadratic polynomial regression model can be expressed as:

[0065] y = β0 + β1x + β2x 2 + ε (3)

[0066] More generally, an nth-order polynomial regression model can be written as:

[0067] y = β0 + β1x + β2x 2 +... + β n x n + ε (4)

[0068] To implement polynomial regression, first, new features need to be constructed from the original data, that is, each independent variable is raised to different powers. For example, if the original data only has x, then x 2 、x 3Such features. These new features, together with the original features, constitute a new design matrix, which serves as the input for the linear regression model.

[0069] The parameters (β0, β1,..., β n ) of the polynomial regression model will, based on this new design matrix, complete parameter identification through linear regression.

[0070] Apply the actual train operation data (time t, speed v(t), train mass m, train braking force F ZD (t)) and the train dynamics model to the constructed polynomial regression algorithm for parameter identification.

[0071] Then the polynomial regression model is changed to:

[0072] Among them,

[0073] Use this polynomial regression model to identify the resistance parameters a, b, c in the virtual formation train dynamics model.

[0074] First, construct the x 2 features in the polynomial regression model, and form a design matrix with the original feature x, expressed as: Then the polynomial regression model can be expressed as:

[0075] Next, complete parameter identification for this regression model through linear regression to obtain the identified parameters a, b, c.

[0076] Finally, calculate y based on the obtained identified parameters a, b, c, which is the identified acceleration a i '(t) of the train. Then the identified speed at any time of the train is v i '(t) = v i (0) + ta i '(t).

[0077] Use the mean square error of speed as the fitness function:

[0078]

[0079] Among them, v i (t) represents the actual speed, and v i '(t) represents the identified speed.

[0080] Calculate the fitness function using the identified train dynamics parameters. Compared with the traditional parameter identification method, the fitness function is smaller and the identification result is more accurate.

[0081] The above step S30 includes: finally, accurately predicting the train braking trajectory based on the train dynamics model with the identification parameters. Figure 3 It is a comparison diagram of the actual braking trajectory and the identified braking trajectory provided by an embodiment of the present invention.

[0082] The train identification speed v i '(t) is obtained based on the train dynamics model with the identification parameters.

[0083] Combined with the train running time t, the s-t diagram of the train identification braking trajectory is predicted.

[0084] The processing flow for generating the braking trajectory of the rear train considering safety constraints based on the braking trajectory of the front train provided by an embodiment of the present invention is as Figure 4 shown, including the following processing procedures:

[0085] Based on the accurate train braking trajectory, the predicted train braking trajectory is used as the braking trajectory of the front train, and further, the braking trajectory of the rear train is generated on the premise of considering safety constraints. First, the predicted train braking trajectory is used as the braking trajectory of the front train to obtain the speed v2'(t) during the braking of the front train and the position of the rear end of the front train at any time s2 (t).

[0086] Then, considering the safety constraint conditions, the braking trajectory of the rear train is dynamically adjusted and generated.

[0087] The safety constraint conditions are:

[0089] In the formula, s2 represents the position of the rear end of the front train at the stop moment; represents the position of the front end of the rear train at the stop moment; s represents the protection distance; s2 (t) represents the position of the rear end of the front train at any time; represents the position of the front end of the rear train at any time; v1'(t) represents the speed during the braking of the rear train; a(t) represents the train acceleration.

[0090] The safety constraint minimizes the train tracking interval while ensuring the safe operation of the train.

[0091] The initial speed v1'(0) of the rear train = v2'(0). If the target stop position is not reached, the speed v1'(t) of the rear train is dynamically adjusted based on the identified speed v2'(t) of the front train in combination with the safety constraints.

[0092] Finally, the speed v1'(t) of the rear train at any time is predicted, and based on this, the predicted braking trajectory of the rear train is generated, obtaining the comparison schematic diagram of the predicted braking trajectories of the front and rear trains in the virtual formation as Figure 5 shown.

[0093] Calculate the initial spacing distance between the front and rear vehicles of the virtual formation based on the predicted braking trajectories of the front and rear vehicles of the virtual formation, and use this initial spacing distance as the minimum tracking spacing of the virtual formation train.

[0094] In summary, the method for generating the minimum tracking spacing of a virtual formation train based on parameter identification proposed by the present invention significantly improves the train operation efficiency and is of great significance to the sustainable development of the urban rail transit system. By collecting the actual operation data of the train, constructing a train dynamics model, and using the polynomial regression algorithm, this method realizes the high-precision prediction of the train braking trajectory, improves the train control accuracy and safety. By improving the train control strategy, the present invention successfully realizes the generation of the minimum tracking spacing of the virtual formation train based on parameter identification. Compared with the method of using a fixed braking rate to achieve the relative braking distance, the train tracking spacing is shortened, and a more efficient and safe train operation is realized.

[0095] The urban rail train under the virtual formation in the embodiment of the present invention can obtain more accurate virtual formation train dynamics parameters through more advanced parameter identification technology. Based on the identified parameter information, the train control strategy can be optimized, the tracking spacing between virtual formation trains can be reduced, and the operation efficiency can be improved.

[0096] Those of ordinary skill in the art can understand that the drawings are only schematic diagrams of an embodiment, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0097] From the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0098] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiments, since they are basically similar to the method embodiments, they are described relatively simply. For the relevant parts, reference can be made to the descriptions in the method embodiments. The device and system embodiments described above are only illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0099] As mentioned above, the above is only the preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for generating the minimum tracking spacing of a virtual marshaling train based on parameter identification, characterized in that: include: Acquiring actual running data of the virtual marshaled train, the actual running data including train running time, train running speed, train mass and train braking force; Considering the train braking force, basic resistance and random interference, a virtual train dynamics model is constructed, and a polynomial regression algorithm is used to identify parameters of the virtual train dynamics model based on the actual operation data of the train; The braking trajectory of the virtual train is predicted based on the dynamic model of the virtual train after parameter identification, and the minimum tracking distance of the virtual train is obtained based on the braking trajectories of the front and rear trains in the virtual train.

2. The method according to claim 1, characterized in that The actual operation data of the virtual marshaled train is obtained, and the actual operation data includes train operation time, train operation speed, train mass and train braking force, including: By collecting the running process of the virtual marshaling train on site, the actual running data of the virtual marshaling train during braking is obtained, and the actual running data includes train running time, train running speed, train mass and train braking force.

3. The method according to claim 2, characterized in that The virtual train dynamics model is constructed by considering the train braking force, basic resistance and random interference, and the parameters of the virtual train dynamics model are identified by using a polynomial regression algorithm based on the actual operation data of the train, including: Considering the braking force, basic resistance and random interference during the train braking process, the dynamic model of the virtual marshaling train is established as follows: In the formula, x i (t) represents the displacement of each train in the virtual marshaling; v i (t) represents the speed of each train in the virtual marshaling; u i (t) represents the control acceleration of the train; F U (t) represents the traction force F on the train QY (t) or braking force F ZD (t); δ i (t) represents the external force per unit mass of each train, including the basic resistance F JB (t) and other additional resistance F related to the environment FJ (t); m represents the mass of the train; g is the acceleration due to gravity; a, b, c are the drag coefficients; The virtual train dynamics model is discretized as follows: Set up an n-order polynomial regression model as: y=β0+β1x+β2x 2 +...+b n x n +e (4) y is the target variable, x is the independent variable, β0 and β1 are model parameters, and ε is the error term; The actual train running data time t, speed v(t), train mass m and train braking force F ZD (t)), and the virtual train dynamics model shown in formula (4) is applied to the polynomial regression algorithm shown in formula (4) for parameter identification; The polynomial regression model is changed to: in, β1=bg,β2=ag,x=v i (t); The polynomial regression model is used to identify the resistance parameters a, b, c in the virtual train dynamics model. Construct the x in the polynomial regression model 2 Features, and together with the original features x, form a design matrix, expressed as: The polynomial regression model can be expressed as: The parameters of the regression model are identified through linear regression, and the identification parameters a, b, and c are obtained; Finally, y is calculated based on the obtained identification parameters a, b, and c, which is the train identification acceleration a. i '(t), then the identified speed of the train at any time is v i '(t)=v i (0)+ta i '(t).

4. The method according to claim 3, characterized in that The method of predicting the braking trajectory of the virtual train based on the dynamic model of the virtual train after parameter identification, and obtaining the minimum tracking distance of the virtual train based on the braking trajectory of the front and rear trains in the virtual train, includes: The train dynamics model based on the identification parameters obtains the train identification speed v i '(t), according to the train identification speed v i The train braking trajectory is predicted based on the predicted braking trajectory of the preceding vehicle. The speed v2'(t) of the preceding vehicle during braking and the position s of the preceding vehicle's rear end at any time are obtained based on the predicted braking trajectory of the preceding vehicle. 2 (t); Set the safety constraints as: In the formula, s 2 Indicates the rear position of the vehicle in front at the time of parking; Indicates the front position of the vehicle behind at the time of parking; s indicates the protection distance; s 2 (t) represents the rear position of the leading vehicle at any time; represents the position of the rear vehicle head at any time; v1'(t) represents the speed of the rear vehicle during braking; a(t) represents the acceleration of the train; The speed of the rear vehicle v1'(t) at any time is predicted based on the identified speed v2'(t) of the leading vehicle combined with safety constraints, and the predicted braking trajectory of the rear vehicle is generated. The initial interval distance between the front and rear vehicles of the virtual formation is calculated based on the predicted braking trajectory of the front and rear vehicles of the virtual formation, and the initial interval distance is used as the minimum tracking distance of the virtual formation train.

Images