Three-sleeper intelligent tamping device and control system suitable for complex line conditions

Abstract

This invention provides a three-sleeper intelligent tamping device and control system suitable for complex track conditions, belonging to the technical field of railway line maintenance and repair equipment. By installing a sleeper empty-sleeper status monitoring device, a pick effective clamping pressure detection device, and a pick clamping angle detection device on the three-sleeper tamping device body, accurate assessment of hidden defects in complex tracks and precise detection of the mechanical state of each pick during tamping operations are achieved. Combined with a ballast stiffness pre-assessment subsystem, an external parameter input / output subsystem, and an effective clamping force self-adjustment subsystem installed on the tamping car body, a three-sleeper intelligent tamping operation control system is constructed. This system enables rapid repair of sleeper empty-sleeper defects and real-time adjustment of tamping operation parameters based on ballast stiffness characteristics and track geometry. It can promptly grasp the ballast condition of complex tracks and scientifically and rationally adjust tamping operation parameters, realizing intelligent maintenance under complex track conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of railway engineering railway line maintenance and repair equipment, in particular to a three-sleeper intelligent tamping device and control system suitable for complex line conditions. The device and control system can quickly detect and eliminate the track sleeper empty suspension in time through intelligent tamping operation, and can comprehensively consider the state of the ballast bed and the geometric shape and position characteristics of the line, adjust the tamping operation parameters in real time, and realize intelligent maintenance and repair of complex line conditions. BACKGROUND

[0002] Under the long-term impact of train loads, the granular ballast bed of the ballast track is prone to uneven settlement, which forces the line geometry to change, resulting in reduced line smoothness, and further affecting train operation safety. In order to maintain the good service state of the line, eliminate the irregularity of the ballast track, and restore the working performance of the ballast bed, large-scale tamping vehicles are often used for line maintenance and repair operations. However, the existing large-scale tamping vehicles cannot detect the state of the ballast bed, and the tamping device can only realize the repeated mechanical tamping operation process, and cannot realize the intelligent operation of "self-learning and self-adjustment" of the tamping parameters according to the state of the ballast bed. This makes the on-site tamping operation mostly rely on experience, and it is difficult to achieve the best effect of the tamping device, and even the line state may become worse after tamping, reducing the mechanical properties of the ballast bed.

[0003] The operating environment of the railway line is complex, and the uneven settlement of the granular ballast bed can cause the track sleeper to be empty and suspended, increase the dynamic response of the track structure, and shorten the service life of the track structure. At present, the detection of the track sleeper empty suspension is mostly based on the "excavation and inspection method", which requires the surrounding ballast to be removed for inspection, which is time-consuming and labor-intensive, and can also disturb the ballast. In addition, manual tamping operation is often used to eliminate the track sleeper empty suspension disease on site, which is slow and difficult to meet the harsh requirements of short maintenance windows and high line state requirements. In addition, due to the complexity of the line state, the sections that need to be tamped may also have track sleeper empty suspension. Under such complex conditions, if blind tamping operation is performed, it may cause the track sleeper empty suspension position to have too much lifting amount, affecting the tamping operation effect. SUMMARY

[0004] The present application aims to provide a three-sleeper intelligent tamping device and control system suitable for complex line conditions, which can accurately evaluate hidden diseases in complex lines and accurately detect the mechanical state of each tamping pick during the tamping operation process, can adjust the tamping operation parameters in real time based on the stiffness characteristics of the ballast bed and the geometric conditions of the line, can timely grasp the state of the ballast bed in complex lines, and can realize intelligent maintenance and repair of complex lines. To solve at least one of the technical problems in the background.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] In one aspect, the present application provides a three-pillow intelligent tamping device, comprising a tamping machine body, a first tamping pick, a second tamping pick and a third tamping pick arranged on the tamping machine body; further comprising:

[0007] A collection device is configured to collect information about the empty lifting state of the bottom of the sleeper, the effective clamping force of each tamping pick during the tamping operation, and the clamping angle.

[0008] A calculation device is connected to the collection device and is configured to calculate the empty lifting height according to the information about the empty lifting state of the bottom of the sleeper, and calculate the ballast bed stiffness according to the effective clamping force of each tamping pick during the tamping operation.

[0009] A processing device is connected to the calculation device and is configured to process the ballast bed stiffness, the line lifting amount and the line shifting amount by using an effective clamping force self-adjusting model, obtain the optimal effective clamping force of the tamping operation, and send an adjustment signal for adjusting the clamping force of the tamping pick to the optimal effective clamping force.

[0010] An adjustment device is connected to the processing device and is configured to receive the adjustment signal and adjust the clamping force of the tamping pick to the optimal effective clamping force during the tamping operation according to the adjustment signal.

[0011] Preferably, the collection device comprises:

[0012] A camera is installed on each tamping pick and is configured to collect information about the empty lifting state of the bottom of the sleeper.

[0013] An embedded pick head pressure sensing film is installed on each tamping pick and is configured to collect the effective clamping force of each tamping pick during the tamping operation.

[0014] A rotation angle measuring device is installed on each tamping pick and is configured to collect the clamping angle of the tamping pick. The rotation angle measuring device is connected to a tension string, the tension string is connected to a fixed wheel, and the fixed wheel is arranged on the pick arm body.

[0015] Preferably, calculating the empty lifting height comprises: performing three-dimensional curved surface fitting and two-dimensional projection on the ballast particle boundary to obtain a jagged boundary, obtaining the distance from each wave valley point to the bottom of the sleeper by using the local maximum height method, and obtaining the maximum empty lifting height of the bottom of the sleeper by means of the global maximum height principle.

[0016] Preferably, calculating the effective clamping force of the tamping pick comprises: collecting the effective clamping pressure of the pick head by the embedded pick head pressure sensing film through the way of partition pressure statistics, and the effective clamping pressure f ij is:

[0017]

[0018] wherein k0 is a pressure conversion coefficient calibrated in the laboratory; p ijis the actual pressure of the embedded pressure sensing film of the i-th row and j-th column of the pick head; A is the area of the region;

[0019] Then, the effective clamping force F of the tamper p is:

[0020] .

[0021] Preferably, assuming that the length of the pull string before the tamper is tamped is L0, the length of the pull string at time t in the clamping stage is L t Then, the effective elongation of the first pull string in the clamping stage is: ; the clamping angle of the tamper is , is the conversion factor between the elongation of the pull string and the clamping angle of the tamper, and the clamping angle of the tamper is converted into the clamping stroke X1 , and L1 represents the effective radius when the tamper rotates.

[0022] Preferably, the effective clamping force of the tamper is a constant value, and the average clamping angle of the two pairs of tamper is calculated according to the clamping angle of each tamper, and the average clamping angle of the intelligent tamper is converted into the stiffness of the ballast bed.

[0023] Preferably, the effective clamping force self-adjusting model is established based on a multi-layer perception machine, the input layer of the effective clamping force self-adjusting model is the initial stiffness of the ballast bed, the track lifting amount and the track shunting amount; the hidden layer adopts a single layer of 6 neurons, and the ReLU function is activated during output; the output layer is the optimal effective clamping force.

[0024] Preferably, the effective clamping force self-adjusting model training and verification includes dividing the data set into a training set, a verification set and a test set, the division ratio is 6:2:2, the loss function adopts MSE, and the optimizer adopts Adam; K-fold cross-validation is used for parameter optimization, the optimal tampering device effective clamping force self-adjusting model is obtained, and the test set is tested; the data set includes multiple groups of data, each group of data includes the stiffness of the ballast bed, the track lifting amount and the track shunting amount, and the corresponding optimal effective clamping force parameter.

[0025] In the second aspect, the application provides a three-tamper intelligent tampering device control method, comprising:

[0026] Collecting the empty hanger state information at the bottom of the sleeper, the effective clamping force of each tamper during the tampering operation, and the clamping angle;

[0027] Calculating the empty hanger height according to the empty hanger state information at the bottom of the sleeper, and calculating the stiffness of the ballast bed according to the effective clamping force of each tamper during the tampering operation and the clamping angle;

[0028] The effective clamping force self-adjusting model is used to process the track bed stiffness, the track lifting amount and the track track shifting amount, so that the optimal effective clamping force of the tamping operation is obtained, and an adjustment signal for adjusting the clamping force of the tamper to the optimal effective clamping force is sent.

[0029] According to the adjustment signal, the clamping force of the tamper during the tamping operation is adjusted to the optimal effective clamping force.

[0030] In a third aspect, the present application provides a three-tie intelligent tamping device control system, comprising:

[0031] The acquisition module is used to acquire the empty tie state information at the bottom of the tie, the effective clamping force of each tamper during the tamping operation, and the clamping angle.

[0032] The calculation module is used to calculate the empty tie height, i.e., the track lifting amount, according to the empty tie state information at the bottom of the tie, and calculate the track bed stiffness according to the effective clamping force of each tamper during the tamping operation and the clamping angle.

[0033] The processing module is used to process the track bed stiffness, the track lifting amount and the track track shifting amount by using the effective clamping force self-adjusting model, so that the optimal effective clamping force of the tamping operation is obtained, and an adjustment signal for adjusting the clamping force of the tamper to the optimal effective clamping force is sent.

[0034] The adjustment module is used to adjust the clamping force of the tamper during the tamping operation to the optimal effective clamping force according to the adjustment signal.

[0035] The present application has the following beneficial effects: By arranging the tie empty lifting state monitoring device, the tamper head effective clamping pressure detection device and the tamper clamping angle detection device on the three-tie tamping device body, the accurate evaluation of the hidden disease of the complex line and the accurate detection of the mechanical state of each tamper during the tamping operation are realized. Combined with the track bed stiffness pre-evaluation subsystem, the external parameter input and output subsystem and the effective clamping force self-adjusting subsystem arranged on the tamping car body, the three-tie intelligent tamping operation control system is constructed, the rapid repair of the tie empty lifting disease and the real-time adjustment of the tamping operation parameters based on the track bed stiffness characteristics and the track geometric conditions are realized. The three-tie intelligent tamping device and the control system can timely grasp the track bed state of the complex line, scientifically and reasonably adjust the tamping operation parameters, and realize the intelligent maintenance of the complex line conditions.

[0036] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following description and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0038] Figure 1 The intelligent tamping device composition and control system block diagram described in the embodiments of the present application.

[0039] Figure 2 The outermost tamping machine structure diagram of the first set of intelligent tamping units described in the embodiments of the present application.

[0040] Figure 3 The micro camera installation position diagram described in the embodiments of the present application.

[0041] Figure 4 The micro camera monitoring area diagram described in the embodiments of the present application. Figure 4 (a) is a front view; Figure 4 (b) is a perspective view.

[0042] Figure 5 The empty lifting height calculation diagram described in the embodiments of the present application.

[0043] Figure 6 The intelligent tamping pick structure diagram described in the embodiments of the present application. Figure 6 (a) is a first tamping pick side view; Figure 6 (b) is a first tamping pick front view; Figure 6 (c) is a third tamping pick front view.

[0044] Figure 7 The effective clamping force calculation principle diagram described in the embodiments of the present application.

[0045] Figure 8 The first tamping pick clamping angle and pick point stroke measurement principle diagram described in the embodiments of the present application.

[0046] Figure 9 The tamping pick-sleeper-ballast bed simulation model diagram described in the embodiments of the present application.

[0047] Figure 10 The effective clamping force self-adjusting model flow chart based on the multi-layer perception machine described in the embodiments of the present application.

[0048] Figure 11 The three-sleeper intelligent tamping device working flow chart described in the embodiments of the present application.

[0049] Wherein, 1-tripod tamper device body; 2-working trolley body; 3-tamper trolley carriage body; 4-first tamper pick; 5-second tamper pick; 6-third tamper pick; 7-first pick arm body; 8-second pick arm body; 9-third pick arm body; 10-first fixed wheel; 11-first pull string; 12-first rotation angle measuring device; 13-second fixed wheel; 14-second pull string; 15-second rotation angle measuring device; 16-third fixed wheel; 17-third pull string; 18-third rotation angle measuring device; 19-micro camera; 20-embedded pick head pressure sensing film; 21-embedded circuit channel; 22-signal transmission line; 23-control system signal transmission line; 24-pillow bottom state display; 25-embankment state pre-evaluation subsystem; 26-external parameter input and output subsystem; 27-effective clamping force self-adjusting subsystem. DETAILED DESCRIPTION

[0050] Embodiments of the present application are described below in detail with reference to the accompanying drawings, wherein like or similar elements are denoted by like or similar reference symbols throughout the drawings. The embodiments described below are examples only, and are not to be construed as limiting the present application.

[0051] It will be understood by those skilled in the art that, as used herein, all terms are intended to have the same meaning as commonly understood by one of ordinary skill in the art in the field of the application, unless explicitly defined otherwise.

[0052] It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined.

[0053] It will be understood by those within the art that, in some aspects of this application, as herein indicated, alternatives to the singular form such as "a", "an" and "the" are contemplated in order to provide for plurals unless otherwise stated. It is further noted that the use of "about" in relation to a given numerical value means ± 10 % of the numeric value unless otherwise stated.

[0054] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, different embodiments or examples described in the specification and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction, if necessary.

[0055] In the description of the specification, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can be explicitly or implicitly included at least one of the features. In the description of the present application, the meaning of "multiple" is two or more than two, unless otherwise specifically limited.

[0056] In the description of the specification, the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present technology and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present technology.

[0057] Unless otherwise specifically defined and limited, the terms "mounting", "connecting", "connecting", "setting" should be broadly understood, for example, it can be fixedly connected, set, or can be detachably connected, set, or integrally connected, set. For those skilled in the art, the specific meaning of the above terms in the present technology can be understood according to the specific circumstances.

[0058] In order to facilitate the understanding of the present application, the present application will be further explained and described in specific embodiments in combination with the drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present application.

[0059] Those skilled in the art should understand that the drawings are only schematic diagrams of the embodiments, and the components in the drawings are not necessarily necessary for the implementation of the present application.

[0060] Embodiment 1

[0061] In this embodiment 1, a three-pillar intelligent tamping device and control system suitable for complex line conditions are provided, which include a tamping machine body, a first tamping pick, a second tamping pick and a third tamping pick arranged on the tamping machine body; further comprising: a collection device for collecting the empty lifting state information of the sleeper bottom, the effective clamping force and the clamping angle of each tamping pick during the tamping operation; a calculation device connected with the collection device, for calculating the empty lifting height according to the empty lifting state information of the sleeper bottom, and calculating the ballast bed stiffness according to the effective clamping force and the clamping angle of each tamping pick during the tamping operation; a processing device connected with the calculation device, for processing the ballast bed stiffness, the line lifting amount and the line shifting amount by using the effective clamping force self-adjusting model to obtain the optimal effective clamping force of the tamping operation, and sending an adjustment signal for adjusting the clamping force of the tamping pick to the optimal effective clamping force; an adjustment device connected with the processing device, for receiving the adjustment signal and adjusting the clamping force of the tamping pick to the optimal effective clamping force during the tamping operation according to the adjustment signal.

[0062] Among them, the collection device includes: a camera mounted on each tamping pick for collecting the empty lifting state information of the sleeper bottom; an embedded pick head pressure sensing film mounted on each tamping pick for collecting the effective clamping force of each tamping pick during the tamping operation; a rotation angle measuring device mounted on each tamping pick for collecting the clamping angle of the tamping pick; the rotation angle measuring device is connected with a tension string, the tension string is connected with a fixed wheel, and the fixed wheel is arranged on the pick arm body.

[0063] Specifically, miniature cameras are installed on the first tamping pick, the second tamping pick and the third tamping pick, then the image processing technology is used to calculate the sleeper empty lifting height, and finally the sleeper bottom state and sleeper empty lifting height information are displayed on the sleeper bottom state display on the tamping car body. In this embodiment, four miniature cameras are arranged in the sleeper end monitoring area, wherein the miniature cameras on the two tamping picks on the same side form a group, constituting an independent monitoring unit. Each independent monitoring unit can calculate the sleeper empty lifting height independently, and the maximum value of the calculation result is used as the evaluation basis. In order to ensure that each independent monitoring unit does not interfere with each other and the imaging effect is best, an invisible area band is artificially set, and the band width is less than the maximum ballast particle size in the ballast bed. The sawtooth boundary is obtained by three-dimensional surface fitting and two-dimensional projection of the ballast particle boundary, the distance from each trough point to the sleeper bottom is obtained by using the local area maximum height method, and the maximum empty lifting height of the sleeper bottom is obtained by means of the global maximum height principle.

[0064] In the embodiment, the embedded pick pressure sensing film installed on each tamper is fixed on the pick of the intelligent tamper by embedding, which can detect the effective force of the pick during the three-pillar tamping operation in real time. Compared with the pick of the traditional tamper, the surface of the pick with the embedded pressure film is not serrated, which reduces the contact area between the pick and the ballast. The embedded pick pressure sensing film realizes the rapid detection of the effective clamping pressure of the pick by the way of partition pressure statistics. Each area is an intelligent sensing element that has been calibrated in the laboratory. The calculation method of the pressure of a single area f ij is shown in formula (1).

[0065] (1)

[0066] In the formula, k0 is the pressure conversion coefficient calibrated in the laboratory; p ij is the actual pressure of the i-th row and j-th column of the embedded pick pressure sensing film; and A is the area of the region.

[0067] Referring to Figure 7 , the actual acting pressure F p of the embedded pick pressure sensing film can be obtained according to formula (2).

[0068] (2)

[0069] Suppose that the length of the pull string before tamping is L0, and the length of the pull string at time t in the clamping stage is L t , then the effective elongation of the first pull string in the clamping stage is : ; the clamping angle of the tamper is , is the conversion coefficient between the elongation of the pull string and the clamping angle of the tamper, the clamping angle of the tamper is converted into the clamping stroke X1, and L1 represents the effective radius when the tamper rotates.

[0070] The effective clamping force of the tamper is a constant value, and according to the obtained clamping angle of each tamper, the average clamping angle of the two pairs of tamper is calculated, and the average clamping angle of the intelligent tamper is converted into the stiffness of the track bed.

[0071] ​​In this embodiment, an effective clamping force self-adjusting model is constructed based on a multi-layer perception mechanism. The input layer of the effective clamping force self-adjusting model is the initial stiffness of the track bed, the track lifting amount, and the track shunting amount. The hidden layer adopts a single layer of 6 neurons, and the ReLU function is activated during output. The output layer is the optimal effective clamping force. The effective clamping force self-adjusting model training and verification includes dividing the data set into a training set, a validation set, and a test set, with a division ratio of 6:2:2. The loss function adopts MSE, and the optimizer adopts Adam. Parameter optimization is performed using K-fold cross-validation to obtain the optimal effective clamping force self-adjusting model of the tamping device, and the test is performed on the test set. The data set includes multiple groups of data, each group of data including the track bed stiffness, the track lifting amount, and the track shunting amount, as well as the corresponding optimal effective clamping force parameters.

[0072] In this embodiment, for the control system of the three sleeper intelligent tamping device described above, the system includes a collection module for collecting the empty hanger state information at the bottom of the sleeper, the effective clamping force of each tamping pick during the tamping operation, and the clamping angle; a calculation module for calculating the empty hanger height according to the empty hanger state information at the bottom of the sleeper, and calculating the track bed stiffness according to the effective clamping force of each tamping pick during the tamping operation and the clamping angle; a processing module for processing the track bed stiffness, the track lifting amount, and the track shunting amount using the effective clamping force self-adjusting model to obtain the optimal effective clamping force of the tamping operation, and issuing an adjustment signal for adjusting the clamping force of the tamping pick to the optimal effective clamping force; and an adjustment module for adjusting the clamping force of the tamping pick during the tamping operation to the optimal effective clamping force according to the adjustment signal. Using the above system, a three sleeper intelligent tamping device control method is realized, including collecting the empty hanger state information at the bottom of the sleeper, the effective clamping force of each tamping pick during the tamping operation, and the clamping angle; calculating the empty hanger height according to the empty hanger state information at the bottom of the sleeper; calculating the track bed stiffness according to the effective clamping force of each tamping pick during the tamping operation and the clamping angle; processing the track bed stiffness, the track lifting amount, and the track shunting amount using the effective clamping force self-adjusting model to obtain the optimal effective clamping force of the tamping operation, and issuing an adjustment signal for adjusting the clamping force of the tamping pick to the optimal effective clamping force; and adjusting the clamping force of the tamping pick during the tamping operation to the optimal effective clamping force according to the adjustment signal.

[0073] Embodiment 2

[0074] In this embodiment 2, a three sleeper intelligent tamping device and control system suitable for complex line conditions are provided. The three sleeper intelligent tamping device is installed with a sleeper empty hanger state monitoring device, a pick effective clamping pressure detection device, and a tamping pick clamping angle detection device on the basis of the original three sleeper tamping device body 1. The three sleeper intelligent tamping control system is provided with a track bed stiffness pre-evaluation subsystem 25, an external parameter input and output subsystem 26, and an effective clamping force self-adjusting subsystem 27 on the original tamping car compartment body 3.

[0075] As Figure 1 shown in the present embodiment 2, the three pillow intelligent tamping device suitable for complex line conditions includes three pillow tamping device body 1, working trolley body 2, tamping trolley compartment body 3, first tamping pick 4, second tamping pick 5, third tamping pick 6, first pick arm body 7, second pick arm body 8, third pick arm body 9, first fixed wheel 10, first pull string 11, first rotation angle measuring device 12, second fixed wheel 13, second pull string 14, second rotation angle measuring device 15, third fixed wheel 16, third pull string 17, third rotation angle measuring device 18, miniature camera 19, embedded pick head pressure sensing film 20, built-in line channel 21, signal transmission line 22, control system signal transmission line 23, pillow bottom state display 24, track bed state pre-evaluation subsystem 25, external parameter input and output subsystem 26, effective clamping force self-adjusting subsystem 27. The three pillow intelligent tamping device is composed of four independent intelligent tamping units, and their working principles are the same. Therefore, it is reasonable to introduce the functional characteristics of the three pillow intelligent tamping device by taking the outermost tamping machine of the first intelligent tamping unit as an example, see Figure 2 .

[0076] In the present embodiment, the sleeper empty lifting state monitoring device realizes the function of collecting sleeper bottom empty lifting state information by the collecting device as described in embodiment 1, including the miniature camera 19 installed on the tamping pick, and the signal transmission line 22 connected with the miniature camera 19, and further including the pillow bottom state display 24 connected with the signal transmission line 22.

[0077] Specifically, as Figure 3 , Figure 4 shown in the present embodiment, the sleeper empty lifting state monitoring device is composed of the miniature camera 19 on the first tamping pick 4, the second tamping pick 5 and the third tamping pick 6, the pillow bottom state display 24 and the signal transmission line 22. By installing the miniature camera 19 on each intelligent tamping pick, then using image processing technology to calculate the sleeper empty lifting height, and finally displaying the sleeper bottom state and sleeper empty lifting height information on the pillow bottom state display 24 on the tamping trolley compartment body 3. Four miniature cameras 19 are arranged in the sleeper end monitoring area, of which the miniature cameras on the two intelligent tamping picks on the same side form a group, constituting an independent monitoring unit. Each independent monitoring unit can calculate the sleeper empty lifting height independently, and take the maximum value of the calculation result as the evaluation basis. In order to ensure that each independent monitoring unit does not interfere with each other and the imaging effect is best, a non-visible area band is artificially set, with a width less than the maximum ballast particle size in the track bed, generally 60mm.

[0078] In this embodiment, the miniature camera 19 of the sleeper empty lifting state monitoring device is installed at a distance of 85 mm from the pick 85 mm to ensure that the miniature camera can clearly observe the ballast at the bottom of the sleeper and the contact state of the sleeper during the tamping stage. Each miniature camera is installed in a wedge-shaped groove, and the maximum length b of the groove is the minimum ballast particle size inside the track bed. As shown in Figure 3 , the slope angle of the groove should be greater than 45° to prevent dust residues from affecting the shooting quality of the miniature camera. As shown in Figure 4 , Figure 5 , when monitoring the sleeper empty lifting state, the width a of the intelligent tamper and the side of the sleeper should be adjusted to between 16 mm and 25 mm, and the minimum distance from the top of the intelligent tamper to the bottom of the sleeper should be 20 mm to provide the best shooting imaging conditions. The miniature camera of the sleeper empty lifting state monitoring device collects images and transmits them to the computer control system in the car body 3 through the signal transmission line 22. The maximum empty lifting height at the bottom of the sleeper is calculated by the calculation module (equivalent to the calculation device in Embodiment 1), including: obtaining a jagged boundary by three-dimensional curved surface fitting and two-dimensional projection of the ballast particle boundary, then obtaining the distance from each valley point to the bottom of the sleeper by using the local area maximum height method, and finally obtaining the maximum empty lifting height d max at the bottom of the sleeper by means of the global maximum height principle. During the sleeper empty lifting state monitoring process, the three-sleeper intelligent tamping device only performs tamping and pick-up operations, and does not need to perform clamping. The time interval between the tamping stage and the pick-up stage is 1 s.

[0079] As shown in Figure 1 , Figure 2 , in this embodiment, the pick head effective clamping pressure detection device, which is equivalent to the collection device, realizes the function of collecting the effective clamping pressure of the pick head in Embodiment 1.

[0080] As shown in Figure 6 , in this embodiment, the pick head effective clamping pressure detection device includes an embedded pick head pressure sensing film 20 on each intelligent tamper and a signal transmission line 22. The pick head pressure sensing film 20 is fixed on the pick head of the intelligent tamper by embedding, which can detect the effective force of the pick head during the three-sleeper tamping operation in real time. Compared with the traditional pick head of the tamper, the surface of the embedded pressure film pick head is not jagged, which reduces the contact area between the pick head and the ballast. To increase the contact area between the pick head of the intelligent tamper and the ballast, the width of the pick head of the intelligent tamper is increased to 150 mm, and the width of the pick head of the half tamper is increased to 110 mm.

[0081] The embedded pick head pressure sensing film 20 realizes rapid detection of the effective clamping pressure of the pick head by means of zoned pressure statistics. Each zone is an intelligent sensing element that has been pressure calibrated in the laboratory. The pressure f ijThe calculation method is shown in formula (1).

[0082] (1)

[0083] In the formula, k0 is a pressure conversion coefficient calibrated in the laboratory, and the value range is 0-10; p ij is the actual pressure of the embedded pickaxe shoe pressure sensing film at the i-th row and j-th column; A is the area of the region.

[0084] As shown in Figure 7 , according to formula (2), the actual acting pressure F p of the embedded pickaxe shoe pressure sensing film can be obtained.

[0085] (2)

[0086] In this embodiment, the clamping angle detection device, which is equivalent to the acquisition device, functions to achieve the function of acquiring the clamping angle on the tamper in the acquisition device in Embodiment 1. As shown in Figure 1 and Figure 8 , the clamping angle detection devices on the first tamper, the second tamper, and the third tamper are of three types, respectively, which are the first tamper clamping angle detection device, the second tamper clamping angle detection device, and the third tamper clamping angle detection device.

[0087] The first tamper clamping angle detection device includes a first fixed wheel 10 installed on the first tamper arm body 7, a first pull string 11 connected with the first fixed wheel 10, and a first rotation angle measurement device 12 connected with the first pull string 11. The second tamper clamping angle detection device includes a second fixed wheel 13 installed on the second tamper arm body 8, a second pull string 14 connected with the second fixed wheel 13, and a second rotation angle measurement device 15 connected with the second pull string 14. The third tamper clamping angle detection device includes a third fixed wheel 16 installed on the third tamper arm body 9, a third pull string 17 connected with the third fixed wheel 16, and a third rotation angle measurement device 18 connected with the third pull string 17.

[0088] In this embodiment, the principles of the above three types of tamper clamping angle detection devices are the same, except that due to the differences in the positions and shapes of the tamper arm bodies, the conversion coefficients between the pull strings and the rotation angles of the intelligent tamper are different.

[0089] In this embodiment, the first tamper clamping angle detection device can be used as an example for installation of other intelligent tamper clamping angle detection devices. Assuming that the length of the first pull string 11 before the first tamper is tamped is L0, and the length of the first pull string 11 at time t in the clamping stage is L t , then the effective elongation of the first pull string 11 in the clamping stage can be obtained through formula (3) Then, the conversion coefficient (0~10) between the elongation of the first pull string 11 calibrated in the laboratory and the clamping angle of the first tamper 4 is used to calculate the clamping angle of the first tamper 4 by formula (4) Meanwhile, the clamping angle of the first tamper 4 is converted into the clamping stroke X1, as shown in formula (5).

[0090] (3)

[0091] (4)

[0092] (5)

[0093] In the formula, L1 is the effective radius of the first tamper 4 when rotating, as shown in Figure 8 .

[0094] In the embodiment, the first tamper 4, the second tamper 5 and the third tamper 6 are fixed on the first tamper arm body 7, the second tamper arm body 8 and the third tamper arm body 9 respectively by bolts. The first tamper 4, the second tamper 5 and the third tamper 6 are basically the same in component, and both are provided with a micro camera 19 and an embedded tamper head pressure sensing film 20 on the tamper body. The difference is that the first tamper 4 and the second tamper 5 are full tamper structures, and the third tamper 6 is a half tamper structure, as shown in Figure 6 .

[0095] In the embodiment, the track bed stiffness pre-evaluation subsystem 25 can realize the pre-detection of the track bed stiffness before the formal tamping operation. The track bed stiffness pre-evaluation subsystem 25, i.e. the virtual function system in the computer control system in the carriage body 3, is equivalent to the calculation module in the embodiment 1, and realizes the function of calculating the track bed stiffness. Specifically as follows:

[0096] (1) The DEM program is used to establish a bulk track bed model, and the sleeper, tamper, tamper arm and other geometric bodies are imported into the MBD program to generate a tamper-sleeper-ballast track bed three-dimensional simulation analysis model, as shown in Figure 9 .

[0097] (2) The three-sleeper tamping operation model of different track bed stiffness is established to analyze the mapping relationship between the track bed stiffness and the average clamping angle of the tamper.

[0098] (3) When detecting the track bed stiffness, the effective clamping force of the tamper is first set as a constant value, then the clamping angle of each intelligent tamper is obtained, and finally the average clamping angle of the two relative tamper pairs is calculated.

[0099] (4) The average clamping angle of the intelligent tamper is converted into the track bed stiffness and transmitted to the external parameter input and output subsystem 26.

[0100] The external parameter input and output subsystem 26 is mainly used for input and output of line condition parameters and ballast bed state parameters, and is a storage medium of initial parameters of tamping operation.

[0101] In the embodiment, the effective clamping force self-adjusting subsystem 27 can quickly predict the optimal effective clamping force parameters of different sleeper empty lifting heights and complex line conditions. The effective clamping force self-adjusting subsystem 27, which is a virtual function system in the computer control system in the car body 3, is equivalent to the processing module and the adjusting module in the embodiment 1, and realizes the functions of calculating the optimal effective clamping force and adjusting the effective clamping force. The self-learning and self-adjusting of tamping operation parameters can be realized. Specifically as follows:

[0102] (1) The internal relationship between the ballast bed stiffness, different track lifting amounts, different track shifting amounts and the optimal clamping force of the intelligent tamper is analyzed by using the three-dimensional simulation analysis model of the tamper-sleeper-ballast bed, and taking the maximum density of the ballast bed under the sleepers after tamping operation and the minimum cumulative settlement rate of the ballast bed under the train load as the evaluation indexes.

[0103] (2) The data obtained from the simulation model is processed, and a data set with the ballast bed stiffness, different track lifting amounts and different track shifting amounts as the independent variables and the optimal clamping force as the target value is established.

[0104] (3) Based on the related theory of multi-layer perception, an effective clamping force self-adjusting model of the tamping device under complex conditions is built. The input layer of the model is the initial stiffness of the ballast bed, the track lifting amount and the track shifting amount; the hidden layer adopts single-layer 6 neurons, and the ReLU function is activated during output. The output layer is the optimal effective clamping force. The data set is divided into a training set, a validation set and a test set, and the division ratio is 6:2:2. The effective clamping force self-adjusting model is trained and verified, the loss function adopts MSE, the optimizer adopts Adam, the initial learning rate is set to 0.0002, and the training times are 5000. The K-fold cross-validation is used for parameter optimization, the optimal effective clamping force self-adjusting model of the tamping device is obtained, and the test is performed on the test set.

[0105] (4) The measured data on site is input into the effective clamping force self-adjusting model of the tamping device, and the robustness and generalization of the effective clamping force self-adjusting model based on multi-layer perception are verified.

[0106] (5) During the tamping operation on site, the initial stiffness of the ballast bed, the track lifting amount and the track shifting amount and other parameters are input into the external parameter input and output subsystem, and the effective clamping force self-adjusting subsystem can quickly predict the optimal clamping force parameters of the tamping operation by using the effective clamping force self-adjusting model based on multi-layer perception.

[0107] Reference Figure 11As shown, the specific working process of the three-bearer intelligent tamping device and the control system is given, facilitating the understanding of the specific implementation process of the three-bearer intelligent tamping device described in the embodiment.

[0108] The three-bearer intelligent tamping device of the embodiment can realize the rapid elimination of hidden diseases of the track support under the coordinated control of the track bed stiffness pre-evaluation subsystem 25, the external parameter input and output subsystem 26, and the effective clamping force self-adjusting subsystem 27. In the process, the track support judged as the empty support does not need to be lifted, the actual empty support height is regarded as the lifting amount in the intelligent tamping operation, and the single-bearer, three-bearer or single-side tamping operation mode is selected according to the empty support area and range. For example, Figure 11 As shown, when the three-bearer intelligent tamping device starts to be used, the miniature camera on each tamper, the tamper pressure detection device and the rotation angle detection device are first turned on. The miniature camera collects the information of the sleeper bottom state displayed in the image and video, the tamper pressure detection device collects the effective clamping force of the tamper, and the rotation angle detection device collects the clamping angle. The collected information of the sleeper bottom state, the effective clamping force of the tamper and the clamping angle are transmitted to the judgment module in the computer control system in the carriage body 3, and the information of the sleeper bottom state is also transmitted to the sleeper bottom state display for display. The judgment module in the computer control system in the carriage body 3 judges whether the track support is in the empty support state according to the information of the sleeper bottom state. If it is in the empty support state, no lifting operation is performed, the empty support condition is analyzed, and the empty support height is calculated by using the calculation device (the virtual function module in the computer control system in the carriage body 3). If the track support is not in the empty support state, the track bed stiffness pre-evaluation subsystem in the computer control system predicts the track bed stiffness according to the collected information of the effective clamping force and the clamping angle of the tamper. The effective clamping force self-adjusting subsystem in the computer control system in the carriage body 3 predicts the optimal effective clamping force according to the calculated empty support height, the initial stiffness of the track bed predicted by the track bed stiffness pre-evaluation subsystem, and the line lifting amount and the line lifting amount input by the external parameter input and output subsystem, and adjusts the clamping force of the tamper until the set effective clamping force is reached.

[0109] In the implementation of the function of the track bed stiffness pre-evaluation subsystem, the relationship between the initial stiffness of the track bed and the average clamping angle is first obtained through modeling analysis, then the clamping angle of each tamper is obtained through the preset effective clamping force, the average clamping angle of the two pairs of tamps is calculated, and finally the average clamping angle is converted into the track bed stiffness. In the implementation of the function of the effective clamping force self-adjusting subsystem, the density of the track bed under the sleeper after tamping is first obtained through modeling analysis, and combined with the cumulative settlement rate of the track bed under the train load after tamping, an effective clamping force self-adjusting model based on a multi-layer perception is constructed. The model processes the calculated empty support height, the predicted initial stiffness of the track bed, and the externally input line lifting amount and line lifting amount to obtain the optimal effective clamping force.

[0110] In summary, the application provides a three-sleeper intelligent tamping device and control system suitable for complex line conditions, by setting a sleeper empty hanging state monitoring device, a pick effective clamping pressure detection device and a tamping pick clamping angle detection device on the three-sleeper tamping device body, accurate evaluation of complex line hidden diseases and accurate detection of the mechanical state of each tamping pick during the tamping operation process are realized. Combined with the ballast bed stiffness pre-evaluation subsystem, the external parameter input and output subsystem and the effective clamping force self-adjusting subsystem set in the tamping car body, a three-sleeper intelligent tamping operation control system is constructed, realizing rapid repair of sleeper empty hanging diseases and real-time adjustment of tamping operation parameters based on the characteristics of the ballast bed stiffness and the line geometric conditions.

[0111] Although the specific embodiments of the application are described above in combination with the drawings, it is not a limitation on the scope of protection of the application, and those skilled in the art should understand that various modifications or changes made by those skilled in the art on the basis of the technical solutions disclosed in the application without creative labor should be covered within the scope of protection of the application.

Claims

1. A three-pillar intelligent tamping device suitable for complex line conditions, comprising a tamping machine body, a first tamper, a second tamper and a third tamper arranged on the tamping machine body; characterized in that, Also comprising: The acquisition device is used for acquiring the sleeper bottom empty lifting state information, the effective clamping force of each tamper in the tamping operation process and the clamping angle; The calculation device is connected with the acquisition device, and is used for calculating the empty lifting height according to the sleeper bottom empty lifting state information; and calculating the ballast bed stiffness according to the effective clamping force of each tamper in the tamping operation process and the clamping angle; The processing device is connected with the calculation device, and is used for processing the ballast bed stiffness, the track lifting amount and the track ballast shifting amount by using the effective clamping force self-adjusting model, obtaining the optimal effective clamping force of the tamping operation, and sending an adjustment signal for adjusting the clamping force of the tamper to the optimal effective clamping force; The adjustment device is connected with the processing device, and is used for receiving the adjustment signal and adjusting the clamping force of the tamper to the optimal effective clamping force during the tamping operation according to the adjustment signal.

2. The three pillow intelligent tamping device suitable for complex line conditions according to claim 1, characterized in that, The acquisition device comprises: A camera mounted on each tamper, which is used for acquiring the sleeper bottom empty lifting state information; An embedded pick head pressure sensing film mounted on each tamper, which is used for acquiring the effective clamping force of each tamper in the tamping operation process; A rotation angle measuring device mounted on each tamper, which is used for acquiring the clamping angle of the tamper; the rotation angle measuring device is connected with a tension string, the tension string is connected with a fixed wheel, and the fixed wheel is arranged on the pick arm body.

3. The three pillow intelligent tamping device suitable for complex line conditions according to claim 1, characterized in that, The calculation of the empty lifting height comprises: obtaining a sawtooth boundary by three-dimensional curved surface fitting and two-dimensional projection of ballast particle boundaries, obtaining the distance from each wave valley point to the sleeper bottom by using a local area maximum height method, and obtaining the maximum empty lifting height of the sleeper bottom by means of a global maximum height principle.

4. The three pillow intelligent tamping device suitable for complex line conditions according to claim 2, characterized in that, The effective clamping force of the ripper is calculated, including the collection of the effective clamping pressure of the ripper shoe through the way of partition pressure statistics of the embedded ripper shoe pressure sensing film, and the single area pressure f ij is: f ij = k0p ij A wherein k0 is the pressure conversion coefficient calibrated in the laboratory; p ij is the actual pressure of the i-th row and j-th column of the embedded pickaxe pressure sensing film; A is the area of the region; Then, the effective clamping force F of the ripper p is:

5. The three pillow intelligent tamping device suitable for complex line conditions according to claim 2, characterized in that, Assuming the length of the pull string before the tamper is not tamped is L0, the length of the pull string at time t in the clamping stage is L t , then the effective elongation of the first pull string in the clamping stage is ΔL t , which is ΔL t =L t -L0; the clamping angle θ of the tamper t is θ t =k1ΔL t , k1 is the conversion coefficient between the elongation of the pull string and the clamping angle of the tamper, and the clamping angle of the tamper is converted into the clamping stroke X1, which is X1=θ t L1, L1 represents the effective radius when the tamper rotates.

6. The three pillow intelligent tamping device suitable for complex line conditions according to claim 1, characterized in that, The effective clamping force of the preset tamper is a constant value, the average clamping angle of the relative two pairs of tamper is calculated according to the obtained clamping angle of each tamper, and the average clamping angle of the tamper is converted into the ballast bed stiffness.

7. The three pillow intelligent tamping device suitable for complex line conditions according to claim 1, characterized in that, The effective clamping force self-adjusting model is constructed based on a multi-layer perception mechanism, the input layer of the effective clamping force self-adjusting model is the initial stiffness of the ballast bed, the track lifting amount and the track ballast shifting amount; the hidden layer adopts a single layer of 6 neurons, and the ReLU function is activated during output; and the output layer is the optimal effective clamping force.

8. The three pillow intelligent tamping device suitable for complex line conditions according to claim 7, characterized in that, The effective clamping force self-adjusting model training and verification comprises: dividing a data set into a training set, a verification set and a test set, the division ratio is 6:2:2, the loss function adopts MSE, and the optimizer adopts Adam; the K-fold cross-validation is used for parameter optimization, the optimal tamping device effective clamping force self-adjusting model is obtained, and the test set is tested; the data set comprises multiple groups of data, each group of data comprises the ballast bed stiffness, the track lifting amount and the track ballast shifting amount, and the corresponding optimal effective clamping force parameter.

9. A control method of a three-pillow intelligent tamping device suitable for complex line conditions, characterized in that, Comprising: Acquiring the sleeper bottom empty lifting state information, the effective clamping force of each tamper in the tamping operation process and the clamping angle; Calculating the empty lifting height according to the sleeper bottom empty lifting state information; and calculating the ballast bed stiffness according to the effective clamping force of each tamper in the tamping operation process and the clamping angle; Processing the ballast bed stiffness, the track lifting amount and the track ballast shifting amount by using the effective clamping force self-adjusting model, obtaining the optimal effective clamping force of the tamping operation, and sending an adjustment signal for adjusting the clamping force of the tamper to the optimal effective clamping force; According to the adjustment signal, the clamping force of the tamper during the tamping operation is adjusted to the optimal effective clamping force.

10. A control system for a three-bench intelligent tamping device suitable for complex track conditions, characterized in that Comprise: The acquisition module is used for acquiring the sleeper bottom empty lifting state information, the effective clamping force of each tamper during the tamping operation and the clamping angle; The calculation module is used for calculating the empty lifting height according to the sleeper bottom empty lifting state information, that is, as the line lifting amount; and calculating the ballast bed stiffness according to the effective clamping force of each tamper during the tamping operation and the clamping angle; The processing module is used for processing the ballast bed stiffness, the line lifting amount and the line lifting amount by using the effective clamping force self-adjusting model, obtaining the optimal effective clamping force of the tamping operation, and issuing an adjustment signal for adjusting the clamping force of the tamper to the optimal effective clamping force; The adjustment module is used for adjusting the clamping force of the tamper during the tamping operation to the optimal effective clamping force according to the adjustment signal.

Images