Large cantilever prestressed concrete cover beam construction method

Abstract

The invention relates to the technical field of large cantilever cover beam construction, and particularly discloses a large cantilever prestressed concrete cover beam construction method, which comprises the following steps: adjusting the height of a sandbox by using a displacement sensor for early warning during construction, monitoring the position of a secondary tensioning platform through the displacement sensor, and adjusting the position of the tensioning platform by using a telescopic arm. And the adjusted platform position is convenient for construction, and the construction efficiency is improved.

Description

Technical Field

[0001] The invention relates to the technical field of large cantilever cap beam construction, and in particular to a large cantilever prestressed concrete cap beam construction method. Background Art

[0002] At present, it is not uncommon for domestic highway bridges to cross major urban roads, and the construction technology of key main projects is relatively mature. However, there are still some unavoidable construction difficulties in urban bridge construction, such as tight construction period, cost control, environmental protection, cross-construction, road maintenance, high-altitude operations, etc. In order to solve the above construction difficulties, various influencing factors should be analyzed before and during construction, key processes should be optimized, and corresponding solutions should be formulated.

[0003] As a transverse beam that supports and covers two or more fulcrums, the cap beam is the focus of urban bridge construction. The cantilever length of the urban bridge cap beam can reach 18 meters, the pier height exceeds 20 meters, and the normal traffic on the road below must be guaranteed during construction, which brings huge challenges to the construction. In addition, when the construction site of the large-span cap beam is limited, the placement of the steel pipe columns in the cap beam support will also be restricted, which also causes inconvenience to the construction. The current secondary tensioning technology of large-span variable-section cap beams has problems such as low efficiency and low intelligence level. Summary of the invention

[0004] The purpose of the present invention is to provide a large cantilever prestressed concrete cap beam construction method to solve the problem that the position of secondary tensioning is difficult to adjust and the efficiency is low during the large cantilever prestressed construction process.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a large cantilever prestressed concrete cap beam construction method, comprising the following steps;

[0006] Step S1, making an enlarged foundation to reduce the bearing capacity requirement of the foundation;

[0007] Step S2, erecting the steel pipe column, using a combination of standard sections and adjustment sections, and connecting the sections with flanges and bolts;

[0008] Step S3, connecting the Bailey pieces using Bailey pins and fixing them laterally using stained glass windows;

[0009] Step S4, cap beam cross slope adjustment, install a displacement sensor at the bottom of the cap beam, transmit the cap beam displacement data to the early warning system, compare it with the preset data, adjust the sandbox height based on the early warning prompt, and complete the cap beam cross slope adjustment;

[0010] Step S5, a steel bar binding platform and temporary protection, using a buckle bracket as temporary protection, the steel bar binding platform is connected and fixed by a cross buckle and the buckle bracket, and the steel bars are positioned according to the cap beam height;

[0011] Step S6, the secondary tensioning platform is installed, the telescopic arm on the secondary tensioning platform is adjusted in position by the telescopic cylinder rod, and the corresponding arm section is locked to realize the telescopic process; a vertical displacement sensor and a horizontal displacement sensor are set on the telescopic arm to monitor the vertical and horizontal displacement differences, and the position of the secondary tensioning platform is adjusted by the telescopic arm.

[0012] Step S7, use the secondary tensioning platform to tension the cap beam.

[0013] As a further technical solution of the above scheme, in step S6, the cap beam is a variable-section cap beam. In combination with the vertical displacement sensor and the horizontal displacement sensor, the coordinates (Xi, Yi, Zi) of the inflection point of the on-site variable-section cap beam and the variable-section function fi (x, y, z) are obtained, and a multi-layer neural network is used to solve the construction position of the secondary tensioning platform.

[0014] As a further technical solution of the above solution, the neural network solving steps are as follows:

[0015] Step S61, constructing a network structure, including an input layer, a hidden layer and an output layer; the input layer is the geometric information of the variable cross-section cap beam obtained by the receiving end, including the inflection point coordinates (Xi, Yi, Zi) and the variable cross-section function f i (x, y, z), the hidden layer is used to extract the features of the input layer, and the output layer is used to output the position coordinates;

[0016] Step S62, weight initialization. Data training can be performed only after the network structure is built. Before training, the weights are initialized with random numbers.

[0017] Step S63, forward propagation, during the training process, the input data is forward propagated through the neural network, in each neuron, the input data domain weight is multiplied, and is processed by the activation function to obtain the output;

[0018] Step S64, calculating the loss, after forward propagation, calculating the gap (i.e., loss) between the output of the neural network and the label;

[0019] Step S65, back propagation, adjusting the weight of the neural network according to the loss to minimize the loss;

[0020] Step S66, weight update, after calculating the derivative of the weight through back propagation, use gradient descent to update the weight;

[0021] Step S67, repeat the training, repeat the above steps S61-S66 until the loss of the neural network converges to the set value.

[0022] As a further technical solution of the above solution, the above neural network propagation operation execution process is as follows:

[0023] i represents the number of layers of the neural network, M i represents the number of neurons in the i-th layer, f i (.) represents the activation function of the neurons in the i-th layer, represents the weight matrix from the i-1th layer to the ith layer, represents the bias from the i-1th layer to the ith layer, represents the net input of the neurons in the i-th layer, represents the output of the neurons in the i-th layer;

[0024] Let a (0) =x, the multi-layer neural network continuously iterates the following formula to propagate information:

[0025] z (i) =W (i) a (i-1) +b (i) ;

[0026] a (i) =f i (z (i) );

[0027] According to the activity value a of the i-1th layer neuron (i-1) Calculate the net activity value z of the neurons in the i-th layer (i) , and then the activity value of the i-th layer neuron is calculated by an activation function. The above two formulas are combined to obtain the following formula:

[0028] z (i) =W (i) f i-1 (z (i-1) )+b (i) ;

[0029] The multi-layer neural network transmits information through neurons layer by layer, and the final output of the network is a (i-1) , the entire network can be viewed as a composite function The vector input x is used as the input a of the first layer (0) , the output a of the i-th layer (i) As the output of the entire function, its parameter passing process is as follows:

[0030]

[0031] Where W and b represent the connection weights and biases of all layers in the multi-layer neural network respectively;

[0032] According to the initial position coordinates of the platform (X 0 , Y 0 , Z 0 ), calculate the platform moving distance (ΔX=XX 0 , ΔY=YY 0 , ΔZ=ZZ 0 ) and direction (the moving direction is determined according to the positive and negative values ​​of ΔX, ΔY and ΔZ); after obtaining the required moving distance and direction of the platform, the telescopic arm is used to adjust the vertical and horizontal position of the platform, and finally the construction position of the secondary tensioning construction platform is obtained.

[0033] Compared with the prior art, the present invention has the following advantages and beneficial effects: the present invention monitors the secondary tensioning platform by arranging vertical displacement sensors and horizontal displacement sensors, and uses telescopic arms to adjust the position of the secondary tensioning platform, thereby adapting to the construction of large cantilever prestressed concrete cap beams and improving construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the installation structure at the sandbox.

[0035] Figure 2 This is a structural schematic diagram of the secondary tensioning platform.

[0036] Figure 3 Flow chart for secondary tensioning platform adjustment.

[0037] The meanings of the numbers in the figure are: sandbox-1; displacement sensor-2; horizontal balancer-3; automatically movable slide pipe-4; automatically adjustable bolt-5; monitoring camera-6; first-stage vertical basic telescopic arm-7; first-stage vertical sliding arm-8; load sensor-9; second-stage vertical basic telescopic arm-10; second-stage vertical sliding arm-11; vertical displacement sensor-12; first-stage horizontal basic telescopic arm-13; first-stage horizontal telescopic arm-14; second-stage horizontal telescopic arm-15; collision sensor-16; horizontal displacement sensor-17; weight adjustment platform-18; moving trolley-19; steel frame-20. DETAILED DESCRIPTION

[0038] The technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention, so as to have a further understanding of the concept of the present invention, the technical problems solved, the technical features constituting the technical scheme and the technical effects brought about.

[0039] like Figure 1 to Figure 3As shown, a large cantilever prestressed concrete cap beam construction method, the technical steps are as follows:

[0040] The first step is to expand the foundation: the large-span variable-section cap beam support with few supports has high requirements for the foundation bearing capacity, so the support foundation is expanded to reduce the foundation bearing capacity requirements. Before pouring concrete, a 1cm thick steel plate is embedded for welding the steel pipe column and the strip foundation;

[0041] The second step is to erect the steel pipe column: the steel pipe column is connected to the pedestal through the flange plate or expansion bolts, and the steel pipe column on the strip foundation is fixed by welding the triangular steel plate to the pre-buried steel plate. The steel pipe column is used in combination with the standard section and the adjustment section, and the sections are connected by flanges and bolts. After the steel pipe column is erected, the sandbox 1 is installed;

[0042] The third step is to install Bailey plates: the main beam uses ordinary 321 Bailey plates, with a total of 6-8 Bailey plates in one beam, connected by Bailey pins; every 5 beams form a group, and two groups are fixed horizontally using stained glass windows;

[0043] Step 4: Adjust the transverse slope of the cap beam: The transverse slope of the large-span cap beam varies between -4% and 4% straight slope, and there is also a 2% herringbone slope. In order to meet all slope requirements, a displacement sensor 2 is installed at the bottom of the cap beam to transmit the displacement data of the cap beam to the early warning system. The early warning system compares the data of the cap beam displacement sensor 2 with the preset slope data, and then prompts the terminal. The terminal can remotely and intelligently adjust the height of the sandbox 1 based on the early warning prompt to complete the adjustment of the transverse slope of the cap beam;

[0044] Step 5: Edge protection and steel bar binding platform: The construction of large-span cap beams is an aerial operation, and a buckle bracket is used as edge protection to ensure construction safety. A50 steel pipes are used to connect and fix the buckle brackets through cross buckles. The steel bars are positioned according to the height of the cap beam, and the steel pipes are used as steel bar skeleton supports;

[0045] Step 6. Installation of the intelligent retractable secondary tensioning platform: the secondary tensioning platform used includes a walking part, a platform part and a load sensor 9, a vertical displacement sensor 12, and a horizontal displacement sensor 17; compared with the existing secondary tensioning platform, the intelligent secondary tensioning platform in the present invention has the advantages of remote control, intelligent adjustment, and high safety. First, the first-level vertical basic telescopic arm 7 is connected to the upper truss through the automatically adjustable bolts 5, the first-level vertical basic telescopic arm 7 is connected to the first-level vertical sliding arm 8, the first-level vertical sliding arm 8 is connected to the secondary tensioning platform, and the second-level vertical basic telescopic arm 10 and the second-level vertical sliding arm 11 are connected below the secondary tensioning platform; when the vertical position needs to be adjusted, the first-level vertical basic telescopic arm 7 and the first-level vertical sliding arm 8 are moved vertically for preliminary adjustment, and then fine-tuned by the second-level vertical basic telescopic arm 10 and the second-level vertical sliding arm 11; in the horizontal direction, the first-level horizontal basic telescopic arm 13, the first-level horizontal telescopic arm 14 and the second-level horizontal telescopic arm 15 are used to adjust the horizontal position, and then a vertical displacement sensor 12 is set on the telescopic arm of the secondary tensioning platform to monitor the vertical position of the secondary tensioning platform. The terminal can calculate the vertical displacement difference according to the reading of the vertical displacement sensor 12 and the preset data, and then adjust the vertical position of the secondary tensioning platform through the automatic first-level vertical sliding arm 8 and the second-level vertical sliding arm 11. A horizontal displacement sensor 17 is set on the telescopic arm of the secondary tensioning platform to monitor the horizontal expansion and contraction of the secondary tensioning platform. The terminal can calculate the horizontal displacement difference based on the reading of the horizontal displacement sensor 17 and the required position for the secondary tensioning, and then adjust the horizontal position of the secondary tensioning platform through the automatic first-level horizontal telescopic arm 14 and the second-level horizontal telescopic arm 15. After the tensioning is completed, the steel frame 20 is driven by the walking part, that is, the moving trolley 19 to move to the next tensioning operation area.

[0046] Combine the vertical displacement sensor 12 and the horizontal displacement sensor 17 to obtain the coordinates (X i , Y i , Z i ) and variable cross-section function f i (x, y, z) is input to the receiving end. After the receiving end obtains the geometric information of the variable-section cap beam, the multi-layer neural network theory is used to solve the platform construction position. The solution process is:

[0047] 1. Construct a network structure, including input layer, hidden layer and output layer. The input layer is the geometric information of the variable cross-section cap beam obtained by the receiving end, including the coordinates of the inflection points (X i , Y i , Z i ) and variable cross-section function f i (x, y, z), the hidden layer is used to extract the features of the input layer, and the output layer is used to output the platform position coordinates.

[0048] 2. Weight initialization: Data training can only be carried out after the network structure is built. Before training, the weights are initialized with random numbers.

[0049] 3. Forward propagation. During the training process, the input data is propagated forward through the neural network. In each neuron, the input data is multiplied by the weight and processed by the activation function to obtain the output.

[0050] 4. Calculate the loss. After forward propagation, calculate the gap between the output of the neural network and the label (i.e., the loss).

[0051] 5. Back propagation, adjust the weights of the neural network according to the loss to minimize the loss.

[0052] 6. Weight update: After calculating the derivative of the weight through back propagation, use gradient descent to update the weight.

[0053] 7. Repeat the training and the above six steps until the loss of the neural network converges to the set value. According to the above process, the coordinates (X, Y, Z) of the platform construction position can be obtained. Based on the above process, the execution process of the required position of the variable cross-section cap beam secondary tensioning platform is as follows:

[0054] i represents the number of layers of the neural network, M i represents the number of neurons in the i-th layer, f i (.) represents the activation function of the neurons in the i-th layer, represents the weight matrix from the i-1th layer to the ith layer, represents the bias from the i-1th layer to the ith layer, represents the net input of the neurons in the i-th layer, Represents the output of the neurons in the i-th layer.

[0055] Let a (0) =x, the multi-layer neural network continuously iterates the following formula to propagate information:

[0056] z (i) =W (i) a (i-1) +b (i) ;

[0057] a (i) =f i (z (i) );

[0058] According to the activity value a of the i-1th layer neuron (i-1) Calculate the net activity value z of the neurons in the i-th layer (i) , and then the activity value of the i-th layer neuron is calculated by an activation function. The above two formulas are combined to obtain the following formula:

[0059] z (i) =W (i) f i-1 (z (i--) )+b (i) .

[0060] The multi-layer neural network transmits information through neurons layer by layer, and the final output of the network is a (i-1) , the entire network can be viewed as a composite function The vector input x is used as the input a of the first layer (0) , the output a of the i-th layer (i) As the output of the entire function, its parameter passing process is as follows:

[0061]

[0062] Among them, W and b represent the connection weights and biases of all layers in the multi-layer neural network respectively.

[0063] According to the initial position coordinates of the platform (X 0 , Y 0 , Z 0 ), calculate the platform moving distance (ΔX=XX 0 , ΔY=YY 0 , ΔZ=ZZ 0) and direction (the moving direction is determined based on the positive and negative values ​​of ΔX, ΔY and ΔZ). After obtaining the required moving distance and direction of the platform, the platform is automatically adjusted in the vertical and horizontal directions, and finally the construction platform required for the secondary tensioning of the variable-section cap beam is obtained. The terminal collects data from the load sensor 9, the vertical displacement sensor 12, and the horizontal displacement sensor 17, including information such as platform height and load, and transmits the collected data to the data center or cloud platform for real-time monitoring, recording, and analysis to identify potential safety risks and operational problems. Based on the results of data anomaly analysis, bottlenecks and unnecessary time-consuming steps in the operation are identified to improve work efficiency and performance. For remote control, through wireless communication technology, the operator remotely controls the extension and retraction of the secondary tensioning platform through a mobile phone App or computer. Remote control is performed through the monitoring camera 6, and the position of the platform can be monitored and adjusted at any time at different locations, thereby improving the convenience and flexibility of operation. At the same time, the intelligent secondary tensioning platform is equipped with an emergency stop button, which is closely connected to the main control system of the platform. Pressing the button immediately cuts off the power supply and stops all movements. In the event of an emergency, the safety of personnel and equipment can be ensured. Finally, collision sensors 16 are set at the key parts of the secondary tensioning platform. Once an abnormal situation is detected, the platform operation will be stopped immediately to avoid accidents. The terminal introduces a fault self-diagnosis function to realize automatic fault detection and alarm to ensure the normal operation of the platform. The intelligent secondary tensioning platform solves the problems of time-consuming manufacturing, installation and disassembly of the existing tensioning platform, welding and transportation difficulties, and can ensure the efficiency of secondary tensioning construction. In addition, a horizontal balancer 3 is provided on the weight adjustment platform 18. When the horizontal balancer 3 is not in the zero position, it will automatically display the deviation and issue an alarm reminder, and automatically adjust the weight size through the weight adjustment platform 18, so that the instrument is adjusted to the zero position, ensuring that the automatically movable slide pipe 4 is in a horizontal state.

[0064] The terms “connection” and “fixation” mentioned in the description of the present invention may refer to fixed connection, processing and forming, welding or mechanical connection. The specific meanings of the above terms in the present invention shall be understood according to the specific circumstances.

[0065] In the description of the present invention, the terms "center", "upper", "lower", "horizontal", "inside", "outside", etc., which indicate the orientation or position relationship, are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying a specific orientation that the device or element referred to must have, and therefore cannot be understood as a limitation on the present invention.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some or all of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a large cantilever prestressed concrete cap beam, characterized in that: The steps include: Step S1, making an enlarged foundation to reduce the bearing capacity requirement of the foundation; Step S2, erecting the steel pipe column, using a combination of standard sections and adjustment sections, and connecting the sections with flanges and bolts; Step S3, connecting the Bailey pieces using Bailey pins and fixing them laterally using stained glass windows; Step S4, cap beam cross slope adjustment, install a displacement sensor at the bottom of the cap beam, transmit the cap beam displacement data to the early warning system, compare it with the preset data, adjust the sandbox height based on the early warning prompt, and complete the cap beam cross slope adjustment; Step S5, a steel bar binding platform and temporary protection, using a buckle bracket as temporary protection, the steel bar binding platform is connected and fixed by a cross buckle and the buckle bracket, and the steel bars are positioned according to the cap beam height; Step S6, the secondary tensioning platform is installed, the telescopic arm on the secondary tensioning platform is adjusted in position by the telescopic cylinder rod, and the corresponding arm section is locked to realize the telescopic process; a vertical displacement sensor and a horizontal displacement sensor are arranged on the telescopic arm to monitor the displacement difference in the vertical direction and the horizontal direction, and the position of the secondary tensioning platform is adjusted by the telescopic arm; Step S7, use the secondary tensioning platform to tension the cap beam.

2. A method for constructing a large cantilever prestressed concrete cap beam as claimed in claim 1, characterized in that: In step S6, the cap beam is a variable cross-section cap beam, and the coordinates (X i , Y i , Z i ) and variable cross-section function f i (x, y, z), and a multi-layer neural network is used to solve the construction position of the secondary tensioning platform.

3. A method for constructing a large cantilever prestressed concrete cap beam as claimed in claim 2, characterized in that: The neural network solving steps are as follows: Step S61, constructing a network structure, including an input layer, a hidden layer and an output layer; the input layer is the geometric information of the variable cross-section cap beam obtained by the receiving end, including the coordinates of the inflection points (X i , Y i , Z i ) and variable cross-section function f i (x, y, z), the hidden layer is used to extract the features of the input layer, and the output layer is used to output the position coordinates; Step S62, weight initialization. Data training can be performed only after the network structure is built. Before training, the weights are initialized with random numbers. Step S63, forward propagation, during the training process, the input data is forward propagated through the neural network, in each neuron, the input data domain weight is multiplied, and is processed by the activation function to obtain the output; Step S64, calculating the loss, after forward propagation, calculating the difference between the output of the neural network and the label; Step S65, back propagation, adjusting the weight of the neural network according to the loss to minimize the loss; Step S66, weight update, after calculating the derivative of the weight through back propagation, use gradient descent to update the weight; Step S67, repeat the training, repeat the above steps S61-S66 until the loss of the neural network converges to the set value.

4. A method for constructing a large cantilever prestressed concrete cap beam as claimed in claim 3, characterized in that: The above neural network propagation operation is performed as follows: i represents the number of layers of the neural network, M i represents the number of neurons in the i-th layer, f i (.) represents the activation function of the neurons in the i-th layer, represents the weight matrix from the i-1th layer to the ith layer, represents the bias from the i-1th layer to the ith layer, represents the net input of the neurons in the i-th layer, represents the output of the neurons in the i-th layer; Let a (0) =c, the multi-layer neural network continuously iterates the following formula to propagate information: z (i) =W (i) a (i-1) +b (i) ; a (i) =f i (z (i) ); According to the activity value a of the i-1th layer neuron (i-1) Calculate the net activity value z of the neurons in the i-th layer (i) , and then the activity value of the i-th layer neuron is calculated by an activation function. The above two formulas are combined to obtain the following formula: z (i) =W (i) f i-1 (z (i-1) )+b (i) ; The multi-layer neural network transmits information through neurons layer by layer, and the final output of the network is a (i-1) , the entire network can be viewed as a composite function The vector input x is used as the input a of the first layer (0) , the output a of the i-th layer (i) As the output of the entire function, its parameter passing process is as follows: Where W and b represent the connection weights and biases of all layers in the multi-layer neural network respectively; According to the initial position coordinates of the platform (X0, Y0, Z0), calculate the platform's moving distance (ΔX=X-X0, ΔY=Y-Y0, ΔZ=Z-Z0) and direction (determine the moving direction based on the positive and negative values ​​of ΔX, ΔY and ΔZ); after obtaining the required moving distance and direction of the platform, use the telescopic arm to adjust the vertical and horizontal position of the platform, and finally obtain the construction position of the secondary tensioning construction platform.

Images