Tunnel lining trolley

Abstract

The present disclosure relates to a tunnel lining trolley, which is applied to the technical field of tunnel trolley technology. It includes: an arch inner template, an arch outer template, a centralized control console, a distance meter, a telescopic rod and a jack. The arch inner template and the arch outer template are connected by a telescopic rod and a jack. The telescopic rod and the jack are connected to the centralized control console. The distance meter is arranged on the arch outer template, including: a distance meter for detecting the distance between the arch outer template and the inner surface of the tunnel, obtaining the distance data and sending it to the centralized control console; a centralized control console for receiving the distance data; inputting the distance data into a pre-trained size prediction neural network to obtain the size data output by the size prediction neural network; when the size data is less than or equal to the first size threshold, controlling the telescopic rod and the jack to telescope based on the size data to control the cross-sectional change of the tunnel lining trolley. The problem of low efficiency in tunnel lining construction is solved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of tunnel trolleys, and in particular to a tunnel lining trolley. Background Art

[0002] With the advancement of science and technology, the requirements for tunnel lining construction are getting higher and higher. Due to the different changes in tunnel cross-section sizes, tunnel variable cross-section lining trolleys are crucial in actual tunnel engineering. Existing tunnel variable cross-section lining trolleys adapt to changes in tunnel cross-section sizes by changing the amount of telescopic rod changes. However, in the process of selecting the amount of telescopic rod changes, small-scale changes are required to meet the lining requirements, or the required specific amount of telescopic rod changes are calculated through complex formulas, which is inefficient and affects tunnel lining construction. Summary of the invention

[0003] In order to solve the above technical problems or at least partially solve the above technical problems, the present disclosure provides a tunnel lining trolley, which can improve the efficiency of tunnel lining construction.

[0004] In order to achieve the above objectives, the technical solutions provided by the embodiments of the present disclosure are as follows:

[0005] The present invention provides a tunnel lining trolley, which includes: an arched inner template, an arched outer template, a centralized control console, a distance meter, a telescopic rod and a jack, wherein the arched inner template and the arched outer template are connected by the telescopic rod and the jack, the telescopic rod and the jack are connected to the centralized control console, and the distance meter is arranged on the arched outer template, including:

[0006] Distance meter, used to detect the distance between the arch outer template and the inner surface of the tunnel, obtain the distance data and send it to the centralized control console;

[0007] A centralized control console is used to receive distance data; input the distance data into a pre-trained size prediction neural network to obtain size data output by the size prediction neural network; wherein the size prediction neural network is trained based on sample data, and the sample data includes multiple groups of sample distance data and sample size data corresponding to the multiple groups of sample distance data; when the size data is less than or equal to a first size threshold, the telescopic rod and the jack are controlled to be telescopic based on the size data to control the cross-sectional change of the tunnel lining trolley.

[0008] As an optional implementation of the embodiment of the present disclosure, the tunnel lining trolley further includes:

[0009] The two ends of the arched inner template and the two ends of the arched outer template are respectively connected to a trolley walking device, and the trolley walking device includes: a rotating disc assembly, a rotating disc upper connecting piece, a rotating disc lower connecting piece and a walking assembly equipped with crawlers, and the walking assembly equipped with crawlers is connected to the rotating disc lower connecting piece; the rotating disc upper connecting piece is used to connect other components;

[0010] The centralized control console is specifically used to: when the size data is greater than a first size threshold, based on the size data, control the rotating disc assembly to change direction, so as to control the walking assembly equipped with crawlers to move along the target direction;

[0011] Wherein, moving along the target direction includes: moving from the arched inner template along a direction perpendicular to the arched axis toward the arched outer template, or moving from the arched outer template along a direction perpendicular to the arched axis toward the arched inner template.

[0012] As an optional implementation of the embodiment of the present disclosure, the tunnel lining trolley further includes:

[0013] The retractable column is connected to a trolley travel device;

[0014] The centralized control console is specifically used to control the telescopic column to extend or shorten according to the size data, so that the third vehicle running device contacts the ground or leaves the ground.

[0015] As an optional implementation of the embodiment of the present disclosure, the telescopic rod includes a main rod and an auxiliary rod;

[0016] The centralized control console is specifically used to: if the size data is less than or equal to the second size threshold, control the main rod to be extended and retracted based on the size data; if the size data is greater than the second size threshold, control the main rod and the auxiliary rod to be extended and retracted based on the size data.

[0017] As an optional implementation of the embodiment of the present disclosure, the centralized control console is also used to: obtain sample data; construct an original neural network; set initial values ​​of multiple parameters in the original neural network, the multiple parameters include: maximum number of iterations, learning accuracy, number of hidden nodes, initial weights, error threshold, initial learning rate; input sample distance data into the original neural network, and output predicted size data; calculate the error of the original neural network based on the predicted size data and the sample size data; if the error is less than or equal to the error threshold, it is determined that the neural network has converged to obtain a pre-trained size prediction neural network.

[0018] As an optional implementation of the embodiment of the present disclosure, the centralized control console is also used to: if the error is greater than the error threshold, correct the initial value to obtain the first neural network; input the sample distance data into the first neural network for training until the error is less than the error threshold.

[0019] As an optional implementation of the embodiment of the present disclosure, the centralized console is also used to: divide the sample data into a training set and a test set; randomly exchange the data included in the training set and the test set to construct a new training set and a test set; and use the new training set and the test set to train the original neural network.

[0020] As an optional implementation of the embodiment of the present disclosure, the centralized control console is also used to: calculate the difference between the predicted size data and the sample size data included in the sample size data; and obtain the error of the original neural network by dividing the difference by the sample size data.

[0021] As an optional implementation of the embodiment of the present disclosure, the tunnel lining trolley further includes:

[0022] A pressure alarm system, wherein the pressure alarm system includes a pressure sensor, and the pressure sensor is arranged on the arch outer template;

[0023] The pressure sensor is used to detect the pressure data of the tunnel top on the arch outer formwork and send it to the centralized control platform;

[0024] The centralized control platform is also used to receive pressure data; determine whether the pressure data is greater than a preset pressure threshold; if the pressure data is greater than the preset pressure threshold, generate a prompt message and send it to the pressure alarm system, and the prompt message is used to prompt to stop pouring.

[0025] As an optional implementation of the embodiment of the present disclosure, the tunnel lining trolley further includes: a crossbeam, a diagonal brace, and a middle support column for supporting the arched inner formwork;

[0026] An outer formwork support assembly, used for supporting the arched outer formwork;

[0027] The detachable widening block is arranged between the arc-shaped inner template and the arc-shaped outer template when the arched inner template is composed of symmetrical arc-shaped inner templates and the arched outer template is composed of symmetrical arc-shaped outer templates.

[0028] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0029] The disclosed embodiment provides a tunnel lining trolley including an arch inner template, an arch outer template, a centralized control console, a distance meter, a telescopic rod and a jack. The arch inner template and the arch outer template are connected by a telescopic rod and a jack, the telescopic rod and the jack are connected to the centralized control console, and the distance meter is arranged on the arch outer template. The distance meter is used to detect the distance between the arch outer template and the inner surface of the tunnel, obtain the distance data and send it to the centralized control console; the centralized control console is used to receive the distance data, and then input the distance data into a pre-trained size prediction neural network to obtain the size data output by the size prediction neural network; when the size data is less than or equal to the first size threshold, the telescopic rod and the jack are controlled to be telescoped based on the size data to control the cross-sectional change of the tunnel lining trolley. Based on the size prediction neural network, the change amount of the telescopic rod is directly obtained through the distance between the arch outer template and the inner surface of the tunnel, so as to realize the automatic change of the cross-sectional change of the tunnel lining trolley, save the time required for multiple adjustments or complex calculations, and thus improve the efficiency of tunnel lining construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0032] Figure 1 A schematic diagram of the structure of a tunnel lining trolley according to an embodiment of the present disclosure Figure 1 ;

[0033] Figure 2 It is a schematic diagram of the layout of the rangefinder according to the embodiment of the present disclosure on the arched outer template;

[0034] Figure 3 A flowchart of a training size prediction neural network according to an embodiment of the present disclosure;

[0035] Figure 4 It is a schematic diagram of the structure of the size prediction neural network described in the embodiment of the present disclosure;

[0036] Figure 5 This is a structural diagram of the telescopic rod described in the embodiment of the present disclosure;

[0037] Figure 6 This is a structural diagram of the trolley walking device according to an embodiment of the present disclosure;

[0038] Figure 7 A schematic diagram of the structure of a tunnel lining trolley according to an embodiment of the present disclosure Figure 2 ;

[0039] Figure 8 A schematic diagram of the structure of a tunnel lining trolley according to an embodiment of the present disclosure Figure 3 . DETAILED DESCRIPTION

[0040] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0042] At present, the variable-section tunnel lining trolley adapts to the changes in tunnel section size by changing the amount of telescopic rod change. However, in the process of selecting the amount of telescopic rod change, small-scale changes are required one at a time to meet the lining requirements, or the required specific amount of telescopic rod change can be accurately calculated through complex formulas, but this will increase the time required for tunnel lining construction and affect construction efficiency.

[0043] In order to solve the above problems, the embodiment of the present disclosure provides a tunnel lining trolley, which can improve the efficiency of tunnel lining construction. The tunnel lining trolley includes an arch inner template, an arch outer template, a centralized control console, a distance meter, a telescopic rod and a jack. Among them, the arch inner template and the arch outer template are connected by a telescopic rod and a jack, the telescopic rod and the jack are connected to the centralized control console, and the distance meter is set on the arch outer template. The distance meter is used to detect the distance between the arch outer template and the inner surface of the tunnel, obtain the distance data and send it to the centralized control console; the centralized control console is used to receive the distance data, and then input the distance data into a pre-trained size prediction neural network to obtain the size data output by the size prediction neural network; when the size data is less than or equal to the first size threshold, the telescopic rod and the jack are controlled to be telescoped based on the size data to control the cross-sectional change of the tunnel lining trolley. It is realized that the change amount of the telescopic rod is directly obtained by the distance between the arch outer template and the inner surface of the tunnel based on the size prediction neural network, which saves the time required for multiple adjustments or complex calculations, thereby improving the efficiency of tunnel lining construction.

[0044] like Figure 1 FIG. 1 is a schematic diagram of a tunnel lining trolley provided in an embodiment of the present disclosure. Figure 1 , Figure 1 The system includes an arch inner template 6, an arch outer template 5, a telescopic rod 4, and a jack 2. It should be noted that the centralized console and the distance meter are Figure 1 It is not shown in the figure, the centralized control console can be arranged at the corresponding working position of the tunnel lining trolley provided by the present disclosure according to actual needs, and the telescopic rod 4 and the jack 2 are both connected to the centralized control console.

[0045] like Figure 1 As shown, the arch inner template 6 is composed of two arc inner templates, which are in symmetrical positions when forming the arch inner template 6, and the arch outer template 5 is composed of four arc outer templates, which are in symmetrical positions when forming the arch outer template 5. The arc inner template and the arc outer template are connected by a telescopic rod 4 and a jack 2. Figure 1 The number of the middle arc inner template, the arc outer template, the telescopic rods and the jacks is shown for example only and is not limited in the present disclosure.

[0046] The distribution of the rangefinder on the surface of the arched outer template 5 is as follows: Figure 2 As shown, Figure 2 The schematic diagram of the layout of the rangefinder on the arch outer template is shown in Figure 1. Taking 18 rangefinders as an example, adjacent rows of laser rangefinders are arranged in a staggered manner, and the distances between laser rangefinders in each row are equal. The present disclosure can use the same arrangement standard on the arch outer templates of tunnel lining trolleys with different cross-sectional sizes.

[0047] like Figure 1 As shown, the tunnel lining trolley provided by the present disclosure includes an arched inner template 6, an arched outer template 5, a telescopic rod 4, a jack 2, a centralized control console, and a distance meter.

[0048] The distance meter is used to detect the distance between the arch outer template and the inner surface of the tunnel, obtain the distance data and send it to the centralized control console. The distance meter includes but is not limited to: a pulse method laser distance meter or a phase method distance meter, which is not limited in the present disclosure. The principle of detecting and obtaining distance data is introduced by taking the phase method distance meter as an example. The phase method distance meter uses the laser beam to perform amplitude modulation and measure the phase delay generated by the modulated light going back and forth once on the measuring line, and then converts the distance data represented by this phase delay according to the wavelength of the modulated light.

[0049] The centralized control console is used to receive the distance data sent by the rangefinder, and then input the distance data into a pre-trained size prediction neural network to obtain the size data output by the size prediction neural network. Further, it is determined whether the size data is less than or equal to a first size threshold. When the size data is less than or equal to the first size threshold, the telescopic rod and the jack are controlled to be telescopic based on the size data to control the cross-sectional change of the tunnel lining trolley. The size prediction neural network is trained based on sample data, and the sample data includes multiple groups of sample distance data and sample size data corresponding to the multiple groups of sample distance data.

[0050] In some embodiments, the centralized control console uses a pre-trained size prediction neural network to obtain the size data required for the tunnel lining trolley to change the cross section. The following will introduce the training process of the size prediction neural network, which can be implemented in the centralized control console and includes S301~S308:

[0051] S301, obtaining sample data.

[0052] In the process of training the dimensional neural network, sample data is first obtained, and the sample data includes sample distance data and sample size data. The disclosed embodiment provides an implementation method for obtaining sample data by means of model testing. Optionally, a proportional model is made by producing tunnel sections and lining trolleys in actual engineering projects. N groups of tests are conducted on the proportional model, and the effect of changing the size of the lining trolley is achieved by changing the length of the telescopic rod. When the size of the lining trolley meets the lining requirements, that is, when the trolley fits the tunnel wall and then performs the lining work, the change in the telescopic rod and the distance data from the outer template of the tunnel lining trolley to the inner surface of the tunnel are recorded. The same method is used to simulate N tunnel tests with different cross-sectional sizes, record the data, and use the distance data from the outer template of the tunnel lining trolley to the inner surface of the tunnel as sample distance data, and the change in the telescopic rod as sample size data, thereby obtaining N groups of sample data.

[0053] Optionally, the sample data also includes tunnel cross-sectional dimensions, which can be calculated based on the above model test. The neural network is trained in combination with the tunnel cross-sectional dimensions, thereby improving the accuracy of the neural network.

[0054] Optionally, the N groups of sample data are normalized to facilitate subsequent training and improve the efficiency of training the neural network.

[0055] Optionally, the above sample data is divided into a training set and a test set. Generally, 80% of the sample data is used as the training set, and the remaining 20% ​​is used as the test set. The data in the training set is used to train the model, and then the error on the test set is used as the generalization error of the final neural network in dealing with real-world scenarios. In the present disclosure, the data in the training set and the test set are stored in groups. For example, a group of sample data includes sample distance data and sample size data. The present disclosure provides an implementation method for randomly exchanging the data included in the training set and the test set to construct a new training set and a new test set. For example, the m groups of sample data included in the training set are exchanged with the m groups of sample data included in the test set, where m is a positive integer. After constructing a new training set and a new test set, the new training set and the test set are used to train the size prediction neural network, so that the trained size prediction neural network has higher accuracy.

[0056] S302, constructing an original neural network.

[0057] The original neural network includes an input layer, a hidden layer, and an output layer. The embodiment of the present disclosure mainly describes the construction of the intermediate layer. The construction of the input layer and the output layer can be specifically referred to the prior art, and the present disclosure will not elaborate on this. The input data of the input layer is the sample distance data, and the output data of the output layer is the predicted size data.

[0058] In the embodiment of the present disclosure, the hidden layer includes: a first unit layer, a second unit layer and a third unit layer. Figure 4 As shown, the first unit layer is used to calculate the horizontal distance and vertical distance between each distance meter and the inner surface of the tunnel; the second unit layer is used to calculate the cross-sectional size of the tunnel; and the third unit layer calculates the change amount of each telescopic rod.

[0059] In some embodiments, each unit layer is provided with a corresponding activation function, and the activation function is selected according to the properties of the output of each unit layer, and the activation function includes a step function, a sigmoid function, a bitangent function, a linear function, etc. This disclosure will not be elaborated here.

[0060] S303, setting initial values ​​of multiple parameters in the original neural network.

[0061] Multiple parameters include: maximum number of iterations, learning accuracy, number of hidden nodes, initial weights, error threshold, and initial learning rate.

[0062] The maximum number of iterations is generally 1000. Since neural network calculations cannot guarantee convergence of iterative results under various parameter configurations, it is necessary to set the maximum number of iterations allowed when the iterative results do not converge.

[0063] Learning accuracy is the ratio of true positive samples to all positive samples in the prediction.

[0064] The number of hidden nodes is selected based on experience, and is generally set to 75% of the number of input layer nodes. For example, if the input layer has 13 nodes and the output layer has 1 node, then the hidden layer can be temporarily set to 10 nodes, which constitutes a 13-10-1 BP neural network model. When training the system, it is actually necessary to compare different numbers of hidden layer nodes, 9, 10, and 11, and finally determine the most reasonable network structure. The number of neuron nodes in the network input layer is the number of characteristic factors (independent variables) of the system, and the number of neuron nodes in the output layer is the number of system targets.

[0065] The initial weights are a set of values ​​that should not be completely equal. Therefore, in the program, a random generator program is used to generate a set of random numbers between -0.5 and +0.5 as the initial weights of the network.

[0066] The threshold is generally set between 0.001 and 0.00001. When the error between two iteration results is less than this value, the system ends the iterative calculation and gives the result.

[0067] The initial learning rate determines the update step size during each iteration of the neural algorithm. It can be adjusted dynamically by multiplying a decay coefficient in each round or according to the change of loss.

[0068] S304, input the sample distance data into the original neural network, and output the predicted size data.

[0069] The distance data included in the sample data is used as the input of the neural network, and the predicted size data can be output based on the above three unit layers, and the size data includes the change amount of the telescopic rod of the tunnel lining trolley.

[0070] In some embodiments, the input layer of the neural network receives the distance data included in the sample data, and passes the distance data to the hidden layer, the neural network performs forward propagation, and the first unit layer included in the hidden layer calculates the horizontal and vertical distances between each laser rangefinder and the inner surface of the tunnel through the distance data input by the input layer; the second unit layer calculates the cross-sectional size of the tunnel through the horizontal and vertical distances between the laser rangefinder and the inner surface of the tunnel obtained in the above-mentioned first unit layer; the third unit layer further calculates the change of the telescopic rod through the cross-sectional size of the tunnel calculated by the above-mentioned second unit layer, and uses this as the predicted size data.

[0071] S305. Calculate the error of the original neural network based on the predicted size data and the sample size data.

[0072] After the hidden layer outputs the predicted size data, the output layer compares the predicted size data with the actual size data. The disclosed embodiment provides an implementation method that calculates the difference between the measured size data and the sample size data, and then divides the difference by the sample size data to calculate the error. After obtaining the error, the neural network performs back propagation, sends the error to all neurons in the hidden layer, and adjusts the weights and thresholds of the neural network. After adjustment, forward propagation is performed again.

[0073] S306: Determine whether the error is less than or equal to an error threshold.

[0074] If the error is greater than the error threshold, execute S307;

[0075] If the error is less than or equal to the error threshold, execute S308.

[0076] S307, modifying the initial value to obtain a first neural network.

[0077] The sample distance data is input into the first neural network for training until the error is less than the error threshold.

[0078] S308: Determine that the neural network converges, and obtain a pre-trained size prediction neural network.

[0079] When the error of the neural network is less than the preset error threshold, it can be determined that the neural network training is complete, and then the test set is used for testing.

[0080] Through the above method, a converged size prediction neural network is trained. In the process of using the size prediction neural network by the centralized control console, the present disclosure inputs the distance data into the size prediction neural network to quickly obtain the size data. Further, it is determined whether the size data is less than or equal to the first size threshold. When the size data is less than or equal to the first size threshold, the telescopic rod and the jack are controlled to telescope based on the size data to control the cross-sectional change of the tunnel lining trolley.

[0081] It is understandable that there is an upper limit to the cross-sectional change of the tunnel lining trolley controlled only by telescopic rods and jacks, that is, the telescopic rods and jacks reach a maximum extension. In practical applications, the maximum extension corresponds to 3 meters, and the first size threshold can be pre-set to 3 meters.

[0082] In some embodiments, the telescopic rod includes a main rod and an auxiliary rod. Figure 5As shown. There are restrictions on the extension and retraction of the main rod 15 and the auxiliary rod 16. A second size threshold can be set in advance. When the size data obtained according to the size prediction neural network is less than or equal to the second size threshold, the centralized control console only controls the main rod to extend and retract to control the area change of the tunnel lining trolley; and when the size data is greater than the second size threshold, it means that only controlling the extension and retraction of the main rod can no longer meet the lining requirements, then the centralized control console controls the main rod and the auxiliary rod to extend and retract at the same time to control the cross-sectional change of the tunnel lining trolley to meet the lining requirements. It can be understood that the second size threshold is less than the first size threshold, and the first size threshold is the value corresponding to when the main rod and the auxiliary rod are fully extended to the upper limit.

[0083] Exemplarily, the second size threshold is set to 2 meters. When the size data is less than 2 meters, the centralized control console only controls the main pole to be extended and retracted; when the size data is greater than 2 meters, the centralized control console controls the main pole and the auxiliary pole to be extended and retracted together.

[0084] The above embodiment realizes directly obtaining the change of the telescopic rod according to the distance data, so that when the size data is less than the first size threshold, the change of the cross-section of the tunnel lining trolley is controlled by the telescopic rod to meet the lining requirements and facilitate the tunnel lining construction.

[0085] When the size data is greater than the first size threshold, the tunnel lining trolley provided by the present disclosure further includes a trolley walking device, Figure 6 This is the structural diagram of the trolley walking device, such as Figure 6 As shown, the trolley travel device includes: a rotating disc assembly 14, a rotating disc upper connecting member 7, a rotating disc lower connecting member 13, and a crawler-equipped travel assembly 8. Among them, the rotating disc upper connecting member 7 is directly connected to the rotating disc assembly 14, the rotating disc assembly 14 is directly connected to the rotating disc lower connecting member 13, and the crawler-equipped travel assembly 8 is connected to the rotating disc lower connecting member 13.

[0086] The connection method of the trolley walking device to the tunnel lining trolley is as follows Figure 8 As shown, the trolley travel device is connected to the arched ends of the arched outer template 5 or the arched inner template 6 by the connecting piece 7 on the rotating disk. The trolley travel device adopts a wheel-track integrated system, which effectively reduces the labor intensity of workers, reduces the number of operators, and saves time for manual track laying.

[0087] In some embodiments, usually, the actual lining requirements cannot be met by only the telescopic rod, and it is necessary to combine the detachable widening block to control the interface change of the tunnel lining trolley. During the installation of the detachable widening block, the size of the trolley movement needs to be adjusted according to the distance data. It is understandable that when the change amount of the telescopic rod has reached the upper limit, that is, the maximum cross-section of the tunnel lining trolley has been reached by the telescopic rod.

[0088] In order to meet the lining requirements, the tunnel trolley needs to be further expanded. First, the various arc-shaped outer template components of the arched outer template move from the arched inner template to the arched outer template along the direction perpendicular to the axis. Correspondingly, the various arc-shaped inner template components of the arched inner template move from the arched inner template to the arched outer template along the direction perpendicular to the axis. Figure 7 As shown, each arc inner template or arc outer template moves in the direction indicated by the double-headed arrow. The movement of the trolley running device connected to the arc inner and outer templates is controlled by the centralized control console.

[0089] The size prediction neural network provided by the embodiment of the present disclosure obtains size data including the size of the trolley movement based on the input distance data. The training process of the size prediction neural network is similar to S301~S308 above, except that the sample data also includes the moving size of the trolley, and an original neural network is constructed, in which the third unit layer of the hidden layer is used to calculate the moving size of the trolley, and then the training can obtain the predicted moving size of the trolley, and further calculate the error between the predicted moving size of the trolley and the moving size of the trolley in the sample data, and train the original neural network until the error is less than the error threshold. Thus, a converged size prediction model is obtained, so that the distance data is input and the moving size of the trolley is output.

[0090] A size prediction neural network is trained in the above manner, and then the distance data is input into the size prediction neural network to obtain size data. When the size data is greater than a first size threshold, the central control console controls the rotating disc assembly in the trolley traveling device to change from moving in a direction parallel to the axis to moving in a direction perpendicular to the axis; and further drives the tracked traveling assembly included in the trolley traveling device to move a corresponding distance in a corresponding direction, so as to change the size of the cross-section of the tunnel lining trolley to meet the lining requirements.

[0091] In some embodiments, the tunnel lining trolley provided by the present disclosure further includes a retractable column, such as Figure 7 As shown, the retractable column 12 is connected to the trolley traveling device. When the centralized control console controls the movement of the trolley traveling device, the centralized control console can also control the retractable column to extend so that the trolley traveling device contacts the ground for movement, thereby changing the cross-sectional size of the tunnel lining trolley, or control the retractable column to shorten so that the trolley traveling device is not in contact with the ground, and the retractable column will serve as a support to ensure the structural stability of the tunnel lining trolley.

[0092] In some embodiments, the tunnel lining trolley further comprises: a cross beam, a diagonal brace, and a middle support column for supporting the arched inner template. Figure 8 As shown in the figure, it includes: a cross beam 9, a diagonal brace 10, and a middle supporting column 3.

[0093] In some embodiments, the process of controlling the movement of the trolley travel device includes controlling the movement of the trolley travel device connected to the arched inner template, controlling the movement of the trolley travel device connected to the arched outer template, and controlling the movement of the trolley travel device connected to the telescopic column. Optionally, the centralized control console moves the telescopic column connected to the trolley travel device to the bottom of the beam, and controls the telescopic column to extend and connect to the beam. The purpose is to leave space for installing the removable widening block after the arc-shaped inner and outer templates at the symmetrical position of the tunnel lining trolley move a corresponding distance, such as Figure 8 As shown, the detachable widening block 1 is arranged between the arched inner template 6 and the arc-shaped outer template 5 to change the cross-sectional size of the tunnel lining trolley to meet the lining requirements.

[0094] The above implementation method controls the change amount of the telescopic rod by a centralized control console to change the cross-sectional size of the tunnel lining trolley, which effectively shortens the time for the trolley to be in place, reduces the construction safety risk, and saves a lot of manpower.

[0095] In some embodiments, the tunnel lining trolley further comprises an outer template support assembly, which is disposed on the inner wall of the arched outer template and is used to support the arched outer template. Figure 8 As shown in the figure, the outer template support assembly 11 is included. The outer template support assembly is composed of a diamond-shaped cross brace and a surrounding belt, and includes a plurality of triangular invariants, so that the longitudinal structure of the trolley is more reasonable.

[0096] In some embodiments, the tunnel lining trolley further includes a pressure alarm system, which includes a pressure sensor, which is arranged on the arch outer formwork to detect the pressure of the tunnel top on the arch outer formwork during the lining process and send it to the centralized control console. When the concrete is conveyed, poured, and lining is performed, the centralized control console turns on the pressure sensor to receive the pressure data, and then determines whether the pressure data is greater than a preset pressure threshold. If the pressure data is greater than the preset pressure threshold, the pressure alarm system will sound an alarm to remind the tunnel top to stop pouring after the pouring work is completed to avoid accidents.

[0097] In summary, the tunnel lining trolley provided by the present disclosure can obtain the change amount of the telescopic rod through the size prediction neural network, so that the telescopic rod can be controlled by the centralized control console to perform corresponding extension and retraction, thereby changing the size of the cross-section of the tunnel lining trolley; or, the size of the trolley movement can be obtained through the size prediction neural network, so that the centralized control console can control the trolley walking device to move in the target direction, which is convenient for the installation of the detachable widening block, thereby changing the size of the cross-section of the tunnel lining trolley to meet the lining requirements. The distance data detected by the rangefinder is combined with the size prediction neural network to directly obtain accurate data on the cross-section change of the trolley, without the need for complex calculations or repeated adjustments, thereby realizing the automation of the cross-section change of the trolley, reducing the consumption of manpower and material resources, saving the time of trolley installation, and thus reducing the cost of tunnel lining construction.

[0098] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0099] The above are only specific embodiments of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A tunnel lining trolley, characterized in that: The invention comprises an arched inner template, an arched outer template, a centralized control console, a rangefinder, a telescopic rod and a jack, wherein the arched inner template and the arched outer template are connected through the telescopic rod and the jack, the telescopic rod and the jack are connected to the centralized control console, and the rangefinder is arranged on the arched outer template, and comprises: The distance meter is used to detect the distance between the arched outer template and the inner surface of the tunnel, obtain distance data and send it to the centralized control console; The centralized control console is used to receive the distance data; Inputting the distance data into a pre-trained size prediction neural network to obtain the size data output by the size prediction neural network; wherein the size prediction neural network is trained based on sample data, and the sample data includes multiple groups of sample distance data and sample size data corresponding to the multiple groups of sample distance data; When the size data is less than or equal to a first size threshold, controlling the telescopic rod and the jack to extend and retract based on the size data to control the cross-sectional change of the tunnel lining trolley; The tunnel lining trolley also includes: two ends of the arched inner template and two ends of the arched outer template are respectively connected to a trolley walking device, the trolley walking device includes: a rotating disc assembly, a rotating disc upper connecting piece, a rotating disc lower connecting piece and a walking assembly equipped with crawlers, the walking assembly equipped with crawlers is connected to the rotating disc lower connecting piece; the rotating disc upper connecting piece is used to connect other components; The centralized control console is specifically used to: when the size data is greater than the first size threshold, based on the size data, control the rotating disc assembly to change direction, so as to control the walking assembly equipped with crawlers to move along a target direction; wherein the movement along the target direction includes: moving from the arch inner template along a direction perpendicular to the arch axis to the arch outer template, or moving from the arch outer template along a direction perpendicular to the arch axis to the arch inner template; The telescopic rod comprises a main rod and an auxiliary rod; The centralized control console is specifically used to: if the size data is less than or equal to a second size threshold, control the main rod to be extended and retracted based on the size data; if the size data is greater than the second size threshold, control the main rod and the auxiliary rod to be extended and retracted based on the size data; The centralized control console is further used to: obtain sample data; construct an original neural network; set initial values ​​of multiple parameters in the original neural network, the multiple parameters including: maximum number of iterations, learning accuracy, number of hidden nodes, initial weights, error threshold, initial learning rate; input the sample distance data into the original neural network, and output the predicted size data; calculate the error of the original neural network according to the predicted size data and the sample size data; if the error is less than or equal to the error threshold, determine that the neural network converges, and obtain the pre-trained size prediction neural network; The tunnel lining trolley also includes: a crossbeam, a diagonal brace, and a middle support column for supporting the arched inner template; An outer formwork support assembly, used for supporting the arched outer formwork; The detachable widening block is arranged between the arc-shaped inner template and the arc-shaped outer template when the arched inner template is composed of symmetrical arc-shaped inner templates and the arched outer template is composed of symmetrical arc-shaped outer templates.

2. The tunnel lining trolley according to claim 1, characterized in that: The tunnel lining trolley further comprises: The retractable column is connected to a trolley travel device; The centralized control console is specifically used to control the retractable column to extend or shorten according to the size data, so that the trolley walking device contacts the ground or leaves the ground.

3. The tunnel lining trolley according to claim 1, characterized in that: The centralized console is also used for: If the error is greater than the error threshold, the initial value is modified to obtain a first neural network; The sample distance data is input into the first neural network for training until the error is less than the error threshold.

4. The tunnel lining trolley according to claim 1, characterized in that: The centralized console is also used for: Dividing the sample data into a training set and a test set; Randomly exchanging the data included in the training set and the test set to construct a new training set and a new test set; The original neural network is trained using the new training set and test set.

5. The tunnel lining trolley according to claim 1, characterized in that: The centralized console is also used for: calculating a difference between the predicted size data and the sample size data included in the sample size data; The error of the original neural network is obtained by dividing the difference by the sample size data.

6. The tunnel lining trolley according to claim 1, characterized in that: The tunnel lining trolley further comprises: A pressure alarm system, wherein the pressure alarm system comprises a pressure sensor, and the pressure sensor is arranged on the arched outer template; The pressure sensor is used to detect the pressure data of the tunnel top on the arched outer template and send it to the centralized control console; The centralized control console is also used to receive the pressure data; Determining whether the pressure data is greater than a preset pressure threshold; If the pressure data is greater than the preset pressure threshold, a prompt message is generated and sent to the pressure alarm system, and the prompt message is used to prompt to stop pouring.

Images