An intelligent reinforcement method, device and system for compression members based on big data

Abstract

The present invention discloses an intelligent reinforcement method, device and system for compression members based on big data. The reinforcement method utilizes big data technology, and based on data such as the temperature, sunlight, environmental humidity, overload information, etc. of the geographical location of the reinforced bridge piers, accurately calculates the required prestress reserve, and automatically formulates a segmented tensioning strategy according to the prestress reserve target, reduces the friction loss, and improves the reinforcement effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and analysis, and specifically relates to an intelligent reinforcement method, device and system for compression members based on big data. Background Art

[0002] Reinforcing bridge piers with prestressed steel wires can prevent cracks from reappearing after repairing and sealing the cracks. However, due to the complex causes of cracks, which are closely related to the ambient temperature, sunlight, humidity, and overload frequency of the concrete pier location, the prestress reserve target is not clear, and due to the frictional resistance between the steel wire and the concrete pier wall, a large prestress loss occurs during the tensioning process, which has a great impact on the reinforcement effect.

[0003] Patent CN115852858A discloses a structural model for reinforcing bridge piers using annular steel wires and proposes a basic reinforcement method framework. CN114808767A discloses a construction method for circumferential prestress of steel wires and improves the tensioning and anchoring device.

[0004] The above patents all relate to the structure of reinforcing concrete bridge piers with prestressed steel wires, but neither clearly defines the prestress reserve target nor proposes a method to reduce frictional resistance loss. Summary of the Invention

[0005] In view of the problems existing in the above-mentioned prior art, the present invention provides an intelligent reinforcement method, device and system for compression members based on big data, which clarifies the prestress reserve when prestressing and reinforcing concrete bridge piers, and automatically formulates a reasonable strategy for reducing frictional resistance loss according to the reinforcement target. The technical solution is as follows:

[0006] In the first aspect, an intelligent reinforcement method for compression members based on big data is provided, which is applied to a cloud server and includes the following steps:

[0007] Receiving a high-definition image of the surface of a concrete bridge pier uploaded by an intelligent handheld terminal, where the compression member is the bridge pier to be reinforced;

[0008] Using a preset image recognition module to identify the friction coefficient between the surface of the bridge pier to be reinforced and the steel wire at preset marking points, generating a friction coefficient matrix, where the preset marking points are arranged at intervals on the steel wire to form a matrix position layout of m rows and n columns;

[0009] Based on the information of sunlight, temperature, humidity, overload magnitude, and overload frequency at the location of the bridge pier to be reinforced, inputting into a trained LSTM neural network model to obtain the cracking stress of the bridge pier to be reinforced ;

[0010] Based on the friction coefficient matrix of the marked points and the cracking stress of the pier to be strengthened, the strengthening plan is calculated, and the strengthening plan includes the determined segmented length of the steel wire rope, the vertical spacing of the steel wire rope, the target friction coefficient of the concrete surface, and the model of the steel wire rope, and the strengthening plan is transmitted to the intelligent handheld terminal.

[0011] In some embodiments, a preset image recognition module is used to recognize the friction coefficient between the surface of the pier to be strengthened and the steel wire rope at the preset marked points, including:

[0012] Collect high-definition images of the surfaces of concrete piers with different roughness levels as training samples, and label the corresponding friction coefficients for the training samples;

[0013] Based on the training samples and the labeled data of the training samples, train the target recognition model, and use the trained target recognition model as the image recognition module;

[0014] For the high-definition image of the surface of the concrete pier to be recognized, based on the image recognition module, obtain the friction coefficient between the surface of the pier to be strengthened and the steel wire rope at the preset marked points in the image.

[0015] In some embodiments, the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency based on the position of the pier to be strengthened is input into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened , including:

[0016] Receive the positioning information uploaded by the intelligent handheld terminal. The positioning information represents the position of the pier to be strengthened, and a high-precision GPS positioning module is installed in the intelligent handheld terminal;

[0017] Based on the position of the pier to be strengthened, obtain the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency throughout the year at the position;

[0018] Input the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency at the position into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened .

[0019] In some embodiments, the training process of the LSTM neural network model includes:

[0020] Using the finite element method, calculate the cracking stress of the pier according to the temperature, sunlight exposure conditions, humidity, overload magnitude, and overload frequency parameters at the pier position;

[0021] Take the temperature, sunlight exposure conditions, humidity, overload magnitude, and overload frequency parameters and the corresponding cracking stress as a training sample;

[0022] Using the temperature, sunlight conditions, humidity, and overloading probability parameters of the training samples as input features, and the pier cracking stress of the training samples as the target value, train the LSTM neural network.

[0023] In some embodiments, the reinforcement plan is based on determined, where ;

[0024] ;

[0025] , where ;

[0026] ;

[0027] The constraint conditions are: ;

[0028] ; ;

[0029] ; ; ;

[0030] Where:

[0031] ——The main cost of the reinforcement plan, in ten thousand yuan;

[0032] ——The cost of reducing the concrete friction coefficient in the reinforcement plan, in ten thousand yuan, is the cost required to reduce the concrete surface friction coefficient within the range of 0.2m×0.1m by 1;

[0033] ——The material and operation cost of the segmented tensioning and anchoring device in the reinforcement plan, in ten thousand yuan, is the material and operation cost of the segmented tensioning and anchoring device generated by splitting the 360° steel wire rope once;

[0034] ——The material and labor cost of the prestressed steel wire rope in the reinforcement plan, in ten thousand yuan, is the material and labor cost of using 1m of prestressed steel wire rope in the reinforcement plan;

[0035] ——The number of points where the friction coefficient needs to be reduced in the friction coefficient matrix. Assume that the number of rows of the points where the friction coefficient needs to be reduced in the friction coefficient matrix is , ;

[0036] —— The coefficient of friction of the marked point where the coefficient of friction needs to be reduced;

[0037] —— The coefficient of friction after the surface treatment of the concrete in the reinforcement plan, that is, the target coefficient of friction;

[0038] —— The cross-sectional diameter of the concrete bridge pier to be reinforced, in m;

[0039] —— The cross-sectional radius of the concrete bridge pier to be reinforced, in m;

[0040] —— The vertical height of the area to be reinforced of the concrete bridge pier, in m;

[0041] —— The vertical spacing of the steel wire ropes in the reinforcement plan, in m;

[0042] —— The vertical distribution quantity of the steel wire ropes in the reinforcement plan;

[0043] —— 1 / 2 of the central angle corresponding to the segmented steel wire rope in the reinforcement plan, in radian system, which is the minimum limit value of the radian corresponding to the segmented steel wire rope;

[0044] —— The segmented length of the steel wire rope in the reinforcement plan, in m;

[0045] —— The minimum compressive stress that can be formed on the surface of the concrete pier column by the reinforcement plan, in MPa;

[0046] —— The tension control stress of the prestressed steel wire rope, in MPa;

[0047] —— The maximum allowable stress of the prestressed steel wire rope, in MPa.

[0048] In some embodiments, the analysis of the reinforcement plan is obtained based on iterative analysis of an intelligent optimization algorithm, and the intelligent optimization algorithm includes: simulated annealing algorithm, particle swarm algorithm, and grey wolf algorithm.

[0049] In a second aspect, an intelligent reinforcement device for compression members based on big data is provided, including:

[0050] An image acquisition unit, configured to receive a high-definition image of the surface of the concrete bridge pier uploaded by an intelligent handheld terminal, where the compression member is the bridge pier to be reinforced;

[0051] The friction coefficient analysis unit is used to identify the friction coefficient between the surface of the pier to be strengthened and the steel wire rope at the preset marked points by using an image recognition module based on a convolutional neural network, and generate a friction coefficient matrix. The preset marked points are arranged at intervals on the steel wire rope, forming a matrix position layout of m rows and n columns.

[0052] The pier cracking stress analysis unit is used to input the information of sunlight, temperature, humidity, overload magnitude, and overload frequency at the position of the pier to be strengthened into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened. ;

[0053] The reinforcement plan analysis unit is used to calculate the reinforcement plan based on the friction coefficient matrix of the marked points and the cracking stress of the pier to be strengthened. The reinforcement plan includes the determined segmented length of the steel wire rope, the vertical spacing of the steel wire rope, the target friction coefficient of the concrete surface, and the steel wire rope model, and transmits the reinforcement plan to the intelligent handheld terminal.

[0054] In a third aspect, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, the steps of the method described in the first aspect above are implemented.

[0055] In a fourth aspect, an intelligent reinforcement system for compression members based on big data is provided, including:

[0056] The segmented prestressed steel wire rope is used to reinforce the compression member, and the compression member is a concrete pier.

[0057] The intelligent segmented tensioning and anchoring module includes a tensioning and anchoring mechanical device and an intelligent control system. The tensioning and anchoring mechanical device is connected to the segmented prestressed steel wire rope, and the intelligent control system is connected to the tensioning and anchoring mechanical device, and is used to perform two-way tensioning and anchoring on the segmented prestressed steel wire rope by controlling the tensioning and anchoring mechanical device.

[0058] The intelligent handheld terminal is communicatively connected to the intelligent control system, and is used to interactively transmit the tensioning and anchoring data to obtain the tensioning force and the anchoring force, and monitor the entire tensioning process; it is also used to take ultra-high pixel photos of the surface of the concrete pier; the intelligent handheld terminal is also communicatively connected to the cloud server, and is used to upload the data collected by the intelligent handheld terminal and receive the data sent by the cloud server.

[0059] The cloud server is communicatively connected to the intelligent handheld terminal, and is used to determine the reinforcement plan according to the method described in the first aspect above based on the data uploaded by the intelligent handheld terminal and send it to the intelligent handheld terminal, and is also used to monitor the entire tensioning process based on the data uploaded by the intelligent handheld terminal.

[0060] In some embodiments, the intelligent handheld terminal includes:

[0061] A high-precision GPS positioning module is used to position the concrete bridge piers to be strengthened;

[0062] An automated modeling module is used to generate a structural and regional scene model of the bridge pier to be strengthened based on the information of the height, diameter, sunlight shielding objects, and material properties of the bridge pier collected on-site.

[0063] An intelligent strengthening method, device, and system for compression members based on big data according to the present invention have the following beneficial effects:

[0064] 1. Based on big data information, the present invention locks parameters such as the temperature, sunlight, humidity, overload magnitude, and frequency at the location of the bridge pier to be strengthened, accurately calculates the cracking load of each concrete bridge pier, and controls the main prestress reserve magnitude. It avoids situations such as excessive or insufficient prestress reserve caused by different cracking loads of bridge piers in different locations, effectively improving the strengthening effect.

[0065] 2. Based on the actual strengthening conditions, the present invention uses an optimization algorithm to determine parameters such as the wire rope model, wire rope segment length, vertical arrangement spacing of wire ropes, and target friction coefficient on the concrete surface based on minimizing the strengthening cost, forming a targeted strengthening plan, implementing a one-pier-column-one-plan approach, and carrying out strengthening work in a targeted manner to maximize the reduction of construction costs.

[0066] 3. The present invention adopts a method of segmentally tensioning and anchoring prestressed wire ropes in combination with concrete surface treatment to reduce the prestress friction loss. Under the condition of taking into account the construction cost, it solves the persistent problems of traditional wire rope strengthening of bridge piers and greatly improves the strengthening efficiency. Description of the Drawings

[0067] Figure 1 is the equipment layout diagram of an intelligent strengthening system for compression members based on big data provided by an embodiment of the present application;

[0068] Figure 2 is the overall structure diagram of the system provided by an embodiment of the present application;

[0069] Figure 3 is the schematic diagram of the LSTM neural network model calculating the cracking stress of bridge pier concrete provided by an embodiment of the present application;

[0070] Figure 4 is the flow schematic diagram of the intelligent strengthening method for compression members based on big data provided by an embodiment of the present application;

[0071] In the figure, 1. Segmented prestressed wire rope; 2. Intelligent handheld terminal; 3. Intelligent segmented tensioning and anchoring module; 4. Cloud server. Detailed Embodiments

[0072] It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0073] An embodiment of the present application provides an intelligent reinforcement system for compression members based on big data, including:

[0074] The segmented prestressed steel wire rope 1 is used to reinforce the compression member, and the compression member is a concrete pier; the specific length of the segmented prestressed steel wire rope 1 is determined according to the reinforcement plan calculated by the cloud server, providing prestress reserve for the concrete to prevent the re-opening of the already closed cracks;

[0075] The intelligent handheld terminal 2 is communicatively connected to the intelligent control system, and is used for interactive transmission of tensioning and anchoring data to obtain the tensile force and anchoring force, and monitor the entire tensioning process; it is also used for taking ultra-high pixel photos of the surface of the concrete pier; the intelligent handheld terminal is also communicatively connected to the cloud server, and is used for uploading the data collected by the intelligent handheld terminal and receiving the data sent by the cloud server. Specifically, the intelligent handheld terminal 2 is equipped with a high-definition panoramic camera to take ultra-high pixel photos of the surface of the concrete pier column, and is equipped with a high-precision GPS positioning module to position the concrete pier to be reinforced. The automatic modeling module inputs information such as the height, diameter, sunlight shielding objects, and material properties of the pier observed on-site, generates a model of the structure and regional scene of the pier to be reinforced, conducts real-time data and image interaction with the cloud server, uploads various information to the cloud server for calculation, and receives the reinforcement plan and calculation cloud map sent by the cloud server to guide on-site construction. It receives the tensile force data transmitted by the intelligent segmented tensioning and anchoring module and monitors the entire tensioning process. In one implementation, after determining the target concrete pier column on-site, the intelligent handheld terminal 2 is used to take high-definition photos of the area to be reinforced, the position information of the concrete pier column is obtained using the high-precision GPS positioning module, the on-site observation data is input, a model of the structure and regional scene of the pier to be reinforced is generated, and all information is uploaded to the cloud server.

[0076] The intelligent segmented tensioning and anchoring module 3 includes a tensioning and anchoring mechanical device and an intelligent control system. The tensioning and anchoring mechanical device is connected to the segmented prestressed steel wire rope, and the intelligent control system is connected to the tensioning and anchoring mechanical device, and is used for bidirectionally tensioning the segmented prestressed steel wire rope by controlling the tensioning and anchoring mechanical device, anchoring with bolts, performing wireless data interaction with the intelligent handheld terminal, receiving the commands of the handheld terminal, and sending the tensioning and anchoring forces to the handheld terminal, etc.;

[0077] The cloud server 4 is communicatively connected to the intelligent handheld terminal, and is used to determine a reinforcement plan based on the data uploaded by the intelligent handheld terminal and send it to the intelligent handheld terminal. It is also used to monitor the entire tensioning process based on the data uploaded by the intelligent handheld terminal. The cloud server has the functions of collecting data from the network, computing and processing, drawing cloud maps, generating reinforcement plans, storing data, and interacting with the intelligent handheld terminal for data.

[0078] The above-mentioned cloud server 4 internally executes an intelligent reinforcement method for compression members based on big data to determine a reinforcement plan and send it to the intelligent handheld terminal to guide on-site construction.

[0079] Specifically, an intelligent reinforcement method for compression members based on big data provided by an embodiment of the present application is applied to a cloud server. The compression member is a concrete pier to be reinforced. The reinforcement method includes the following steps:

[0080] Step 1, receive high-definition images of the surface of the concrete pier uploaded by the intelligent handheld terminal;

[0081] Step 2, use a preset image recognition module to recognize the friction coefficient between the surface of the pier to be reinforced and the steel wire rope at preset marking points, and generate a friction coefficient matrix;

[0082] It can be understood that multiple prestressed steel wire ropes are arranged at different vertical height positions of the pier. A marking point is set at every preset length (for example, 10 cm) on each steel wire rope. Assume that the number of vertical recognition points on the surface of the area to be reinforced of the concrete pier (that is, the number of steel wire ropes at different vertical height positions) is m, and the number of horizontal recognition points on the surface of the area to be reinforced of the concrete pier (that is, the number of recognition points on each steel wire rope) is n, forming a marking point position distribution of m rows and n columns, and correspondingly forming a friction coefficient matrix of m rows and n columns.

[0083] The cloud server, according to the high-definition images of the concrete surface uploaded by the intelligent handheld device, through an embedded image recognition module based on a convolutional neural network, and the image recognition module is a fine-grained image recognition module, recognizes the distribution of the friction coefficient between the surface of the concrete to be reinforced and the steel wire rope, the distance between recognition points is 10 cm, generates a friction coefficient matrix and generates a three-dimensional cloud map.

[0084] Specifically, for step 2, using a preset image recognition module to recognize the friction coefficient between the surface of the pier to be reinforced and the steel wire rope at preset marking points includes:

[0085] Step 21, collect high-definition images of the surfaces of concrete piers with different roughness levels as training samples, and label the corresponding friction coefficients for the training samples;

[0086] Step 22: Train the target recognition model based on the training samples and the labeled data of the training samples. Use the trained target recognition model as the image recognition module. The target recognition model is constructed based on a convolutional neural network. In one implementation, the target recognition model uses the YOLO v5 model;

[0087] Step 23: For the high-definition image of the surface of the concrete pier to be recognized, based on the image recognition module, obtain the friction coefficient between the surface of the pier to be strengthened and the steel wire rope at the preset marking points in the image.

[0088] It can be understood that the friction coefficient characterizes the roughness of the concrete surface. There are obvious differences in the images of concrete surfaces with different roughness levels, which can be recognized by the target recognition model. In Steps 21 - 23, a large number of high-definition pictures of the pier surface are taken in the early stage, and the corresponding friction coefficient μ is obtained through traditional methods. Use the high-definition pictures of the pier surface and the corresponding friction coefficient μ as the training set and the test set to train the YOLO v5 model. Use the high-definition pictures of the concrete surface as the input and the friction coefficient μ as the output to obtain the target recognition model applicable to the embodiments of the present application. Embed the trained model into the cloud server to obtain the friction coefficient through the collected images.

[0089] Step 3: Input the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency of the position of the pier to be strengthened into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened ;

[0090] In the embodiments of the present application, based on big data information, lock parameters such as temperature, sunlight exposure, humidity, overload magnitude, and frequency of the position of the pier to be strengthened, accurately calculate the cracking load of each concrete pier, and control the main reinforcement prestress reserve magnitude. Avoid situations such as excessive or insufficient prestress reserve caused by different cracking loads of piers in different locations, effectively improving the reinforcement effect.

[0091] Specifically, this Step 3 includes the following steps:

[0092] Step 31: Receive the positioning information uploaded by the intelligent handheld terminal. The positioning information represents the position of the pier to be strengthened;

[0093] Step 32: Based on the position of the pier to be strengthened, obtain the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency of this position throughout the year;

[0094] Step 33: Input the information of sunlight exposure, temperature, humidity, overload magnitude, and overload frequency of this position into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened .

[0095] In the embodiment of the present application, the cloud server automatically collects the annual sunshine, temperature, humidity, overload magnitude, and overload frequency information at the location according to the actual positioning information of the bridge pier to be strengthened uploaded by the intelligent handheld device, and combines the structure and regional scene model of the bridge pier to be strengthened uploaded by the intelligent handheld device. Through the LSTM neural network model completed by embedded training, the cracking stress of the concrete bridge pier is calculated.

[0096] In the above step 33, during the training process of the LSTM neural network model, it includes:

[0097] Step 331, using the finite element method, calculate the cracking stress of the bridge pier according to the parameters of temperature, sunshine condition, humidity, overload magnitude, and overload frequency;

[0098] Step 332, taking the parameters of temperature, sunshine condition, humidity, overload magnitude, and overload frequency and the corresponding cracking stress as a training sample;

[0099] Step 333, based on the parameters of temperature, sunshine condition, humidity, and overload occurrence probability of the training sample as input features, and the cracking stress of the bridge pier in the training sample as the target value, train the LSTM neural network.

[0100] In the embodiment of the present application, first use the finite element method to calculate the corresponding cracking stress results of the bridge pier according to a large number of different parameters such as temperature, sunshine condition, humidity, overload magnitude frequency, etc. Then, take the temperature, sunshine condition, humidity, and overload magnitude frequency as feature values, and the cracking stress of the bridge pier as the target value, sort out the training set and test set, and use the training set and test set data to train the LSTM neural network model. When the loss function value drops to the minimum and stabilizes, the training of the LSTM neural network model is completed.

[0101] Step 4, calculate the reinforcement plan based on the friction coefficient matrix of the marked points and the cracking stress of the bridge pier to be strengthened. The reinforcement plan includes the determined wire rope segment length, wire rope vertical spacing, target friction coefficient of the concrete surface, and wire rope model, and transmit the reinforcement plan to the intelligent handheld terminal.

[0102] In the embodiment of the present application, the cloud server can design the reinforcement plan according to the friction coefficient matrix and the cracking stress of the concrete bridge pier to be strengthened. Combining wire rope segmentation, concrete surface friction coefficient treatment, and wire rope spacing setting, on the basis of meeting the reinforcement requirements, adopt the most economical reinforcement plan.

[0103] The reinforcement plan is based on determined, where ;

[0104] ;

[0105] , where ;

[0106] ;

[0107] The constraint conditions are: ;

[0108] ; ;

[0109] ; ; ;

[0110] Among them:

[0111] ——The main cost of the reinforcement plan, in ten thousand yuan;

[0112] ——The cost of reducing the concrete friction coefficient in the reinforcement plan, in ten thousand yuan, is the cost required to reduce the concrete surface friction coefficient by 1 within the range of 0.2m×0.1m in area (0.2m is the necessary vertical range for the tensioned steel wire rope), The specific value of can be 1.24;

[0113] ——The material and operation cost of the segmented tensioning and anchoring device in the reinforcement plan (i.e., the intelligent segmented tensioning and anchoring module in the embodiments of the present application), in ten thousand yuan, is the material and operation cost of the segmented tensioning and anchoring device (i.e., the intelligent segmented tensioning and anchoring module in the embodiments of the present application) generated each time the 360° steel wire rope is split, The specific value of can refer to the average cost in today's market, including labor, materials and other costs; the subsequent calculation part represents the total number of segments of the 360° steel wire rope split; for example, in one estimation method,

[0114] ——The material and labor cost of the prestressed steel wire rope in the reinforcement plan, in ten thousand yuan, is the material and labor cost of each 1m of prestressed steel wire rope used in the reinforcement plan, The specific value of The part is the total length of the steel wire rope used in the whole reinforcement; for example, in one estimation method, It can be 1.3;

[0115] —— The number of points where the friction coefficient needs to be reduced in the friction coefficient matrix. Assume that the number of rows of the points where the friction coefficient needs to be reduced in the friction coefficient matrix is , ;

[0116] —— The friction coefficient of the marked points where the friction coefficient needs to be reduced;

[0117] —— The friction coefficient after the surface treatment of the concrete in the reinforcement scheme, that is, the target friction coefficient;

[0118] —— The cross-sectional diameter of the concrete pier to be reinforced, in m;

[0119] —— The cross-sectional radius of the concrete pier to be reinforced, in m;

[0120] —— The vertical height of the area to be reinforced of the concrete pier, in m;

[0121] —— The vertical spacing of the steel wire ropes in the reinforcement scheme, in m;

[0122] —— The vertical distribution quantity of the steel wire ropes in the reinforcement scheme;

[0123] —— 1 / 2 of the central angle corresponding to the steel wire rope after the reinforcement scheme is segmented, in radian system, is the minimum limit value of the radian corresponding to the segmented steel wire rope. Although the reinforcement effect can be improved by segmenting the steel wire rope, there is a minimum limit for the segmented length of the steel wire rope. According to research, the minimum corresponding radian after the steel wire rope is segmented should not be less than 0.056. That is, in one implementation method, = 0.056;

[0124] —— The segmented length of the steel wire rope in the reinforcement scheme, in m;

[0125] —— The minimum compressive stress that can be formed on the surface of the concrete pier column in the reinforcement scheme, in MPa;

[0126] —— The tension control stress of the prestressed steel wire rope, in MPa;

[0127] ——The maximum allowable stress of the prestressed steel wire rope, in MPa.

[0128] In the above formula, W1, W2, and W3 respectively represent the cost of reducing the concrete friction coefficient in the reinforcement plan, the material and operation cost of the intelligent segmented tensioning and anchoring module in the reinforcement plan, and the material and labor cost of the prestressed steel wire rope in the reinforcement plan. The relationship among these three costs is that when one increases, the others decrease. Taking the sum of the three costs as the objective function, an optimization algorithm is used to calculate the plan with the lowest total cost, and obtain the friction coefficient of the concrete surface after treatment in the reinforcement plan with the lowest total cost , the vertical spacing h of the steel wire ropes in the reinforcement plan, and the segmented length of the steel wire ropes , the tension control stress of the prestressed steel wire rope . Based on the required tension control stress of the prestressed steel wire rope determine the maximum allowable stress of the prestressed steel wire rope , so as to determine the steel wire rope model that meets the tension control stress.

[0129] In the cloud server, intelligent optimization algorithms such as simulated annealing algorithm, particle swarm algorithm, and grey wolf algorithm can be used to solve the above formula to determine l, μ s , h, σ max and other parameters, that is, determine the steel wire rope model, the segmented length of the steel wire rope, the vertical spacing of the steel wire ropes, the target friction coefficient of the concrete surface, etc., form a reinforcement plan, and transmit it to the intelligent handheld terminal to guide the on-site construction.

[0130] Based on the actual reinforcement conditions, the embodiments of the present application use an optimization algorithm based on minimizing the reinforcement cost to determine parameters such as the steel wire rope model, the segmented length of the steel wire rope, the vertical layout spacing of the steel wire ropes, and the target friction coefficient of the concrete surface, form a targeted reinforcement plan, achieve one plan for each pier column, and carry out the reinforcement work in a targeted manner, maximizing the reduction of construction costs.

[0131] The embodiments of the present application adopt the method of segmentally tensioning and anchoring the prestressed steel wire rope combined with the concrete surface treatment to reduce the prestress friction loss. Under the condition of taking into account the construction cost, it solves the stubborn problems of the traditional steel wire rope for reinforcing pier columns and greatly improves the reinforcement efficiency.

[0132] Based on the above intelligent reinforcement method for compression members based on big data, the embodiments of the present application provide an intelligent reinforcement device for compression members based on big data, including:

[0133] An image acquisition unit, configured to receive a high-definition image of the concrete pier surface uploaded by the intelligent handheld terminal. The image is an image of the compression member during the reinforcement process, and the compression member is the pier to be reinforced;

[0134] The friction coefficient analysis unit is used to identify the friction coefficient between the surface of the pier to be strengthened and the steel wire rope at the preset marking points by using an image recognition module based on a convolutional neural network, and generate a friction coefficient matrix. The preset marking points are arranged at intervals on the steel wire rope, forming a matrix position layout of m rows and n columns.

[0135] The pier cracking stress analysis unit is used to input the information of sunlight, temperature, humidity, overload level, and overload frequency at the position of the pier to be strengthened into the trained LSTM neural network model to obtain the cracking stress of the pier to be strengthened. ;

[0136] The reinforcement plan analysis unit is used to calculate the reinforcement plan based on the friction coefficient matrix of the marking points and the cracking stress of the pier to be strengthened. The reinforcement plan includes the determined segment length of the steel wire rope, the vertical spacing of the steel wire rope, the target friction coefficient of the concrete surface, and the steel wire rope model, and transmits the reinforcement plan to the intelligent handheld terminal.

[0137] For the specific limitations of the intelligent reinforcement device for compression members based on big data, reference can be made to the limitations of the intelligent reinforcement method for compression members based on big data in the above text, which will not be elaborated here. Each unit in the above intelligent reinforcement device for compression members based on big data can be implemented in whole or in part by software, hardware, and their combination. The above units can be embedded in the processor in the cloud server 4 in the form of hardware or independent of it, or stored in the memory in the cloud server 4 in the form of software, so that the processor can call and execute the operations corresponding to the above units.

[0138] As an example, the intelligent reinforcement device for compression members based on big data provided by the embodiments of the present application can be directly embodied as a combination of software modules executed by a processor. The software modules can be located in a storage medium, and the storage medium is located in the memory. The processor reads the executable instructions included in the software modules in the memory and combines the necessary hardware (for example, including the processor and other components connected to the bus) to complete the intelligent reinforcement method for compression members based on big data provided by the embodiments of the present invention.

[0139] The embodiments of the present application also provide a computer-readable storage medium, on which computer instructions are stored. When the instructions are executed by a processor, the steps of the above intelligent reinforcement method for compression members based on big data are implemented. The computer-readable storage medium includes: electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, and any suitable combination of the above. It can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage nodes, etc.

[0140] The present invention is not limited to the above specific embodiments. Those of ordinary skill in the art, starting from the above concepts and without creative efforts, can make various changes, all of which fall within the protection scope of the present invention.

Claims

1. A method for intelligent reinforcement of compression components based on big data, applied to cloud servers, characterized in that: The steps include: Receiving a high-definition image of a concrete bridge pier surface uploaded by a smart handheld terminal, wherein the compression member is a bridge pier to be reinforced; Using a preset image recognition module, the friction coefficient between the surface of the bridge pier to be reinforced and the steel wire rope at the preset marking points is identified to generate a friction coefficient matrix, wherein the preset marking points are arranged at intervals on the steel wire rope to form a matrix position layout with m rows and n columns; The cracking stress of the bridge pier to be reinforced is obtained by inputting the trained LSTM neural network model based on the sunshine, temperature, humidity, overload level and overload frequency information of the bridge pier to be reinforced. ; A reinforcement scheme is obtained based on the friction coefficient matrix of the marked points and the cracking stress calculation of the bridge pier to be reinforced. The reinforcement scheme includes the determined segment length of the wire rope, the vertical spacing of the wire rope, the target friction coefficient of the concrete surface, and the wire rope model. The reinforcement scheme is transmitted to the smart handheld terminal; The reinforcement scheme is determined based on minW, where: ; ； ； ； The constraints are: ; ； ； in: W——Main cost of reinforcement plan, in ten thousand yuan; ——The cost of reducing the friction coefficient of concrete in the reinforcement scheme, in ten thousand yuan, The cost required to reduce the friction coefficient of the concrete surface by 1 in an area of ​​0.2m×0.1m; ——Material and operating costs of the segmented tension anchorage devices in the reinforcement scheme, in ten thousand yuan, The material and operating costs of the segmented tensioning anchorage device generated each time the 360° wire rope is split; ——Material and labor costs of prestressed steel wire ropes in the reinforcement scheme, in ten thousand yuan, The material and labor cost for each 1m of prestressed steel wire rope used in the reinforcement scheme; p——The number of points in the friction coefficient matrix where the friction coefficient needs to be reduced. Assume that the number of rows of points in the friction coefficient matrix where the friction coefficient needs to be reduced is ; - The friction coefficient of the marking point where the friction coefficient needs to be reduced; ——The friction coefficient of the concrete surface after reinforcement scheme treatment, that is, the target friction coefficient; D——diameter of the cross section of the concrete pier to be reinforced, in meters; R——radius of the cross section of the concrete pier to be reinforced, in m; H——vertical height of the area to be reinforced of the concrete pier, in meters; h——vertical spacing of steel wire ropes in the reinforcement scheme, in m; a——the number of vertically distributed steel wire ropes in the reinforcement scheme; ——The segmented steel wire rope in the reinforcement scheme corresponds to 1 / 2 of the central angle of the circle, in radians. It is the minimum limit value of the arc corresponding to the wire rope segmentation; l——the length of the steel wire rope segment in the reinforcement scheme, in meters; σ——the minimum compressive stress that can be formed on the surface of the concrete pier by the reinforcement scheme, in MPa; ——Prestressed steel wire rope tensioning control stress, unit is MPa; ——The maximum allowable stress of prestressed steel wire rope, in MPa.

2. According to the big data-based intelligent reinforcement method for compressive components of claim 1, it is characterized in that: The preset image recognition module is used to identify the friction coefficient between the surface of the bridge pier to be reinforced and the steel wire rope at the preset mark point, including: Collect high-definition images of concrete pier surfaces with different degrees of roughness as training samples, and annotate the corresponding friction coefficients for the training samples; Training a target recognition model based on the training samples and the labeled data of the training samples, and using the trained target recognition model as the image recognition module; For a high-definition image of the concrete pier surface to be identified, the friction coefficient between the surface of the pier to be reinforced and the steel wire rope at a preset marking point in the image is obtained based on the image recognition module.

3. The intelligent reinforcement method for compressive components based on big data according to claim 1 is characterized in that: The information of sunshine, temperature, humidity, overload level and overload frequency at the location of the bridge pier to be reinforced is input into the trained LSTM neural network model to obtain the cracking stress σ_0 of the bridge pier to be reinforced, including: Receiving positioning information uploaded by a smart handheld terminal, wherein the positioning information represents the position of the bridge pier to be reinforced, and the smart handheld terminal is equipped with a high-precision GPS positioning module; Based on the location of the bridge pier to be reinforced, obtain the annual sunshine, temperature, humidity, overload level and overload frequency information of the location; The acquired sunshine, temperature, humidity, overload level, and overload frequency information are input into the trained LSTM neural network model to obtain the cracking stress of the bridge pier to be reinforced. .

4. The intelligent reinforcement method for compressive components based on big data according to claim 1 is characterized in that: The training process of the LSTM neural network model includes: The finite element method is used to calculate the cracking stress of the bridge pier according to the temperature, sunshine conditions, humidity, overload level and overload frequency parameters at the bridge pier location. The temperature, sunshine conditions, humidity, overload magnitude, overload frequency parameters and the corresponding cracking stress are used as a training sample; Based on the temperature, sunshine conditions, humidity, and overload probability parameters of the training samples as input features, the pier cracking stress of the training samples is used as the target value to train the LSTM neural network.

5. The intelligent reinforcement method for compressive components based on big data according to claim 1 is characterized in that: The analysis of the reinforcement scheme is obtained based on iterative analysis of an intelligent optimization algorithm, and the intelligent optimization algorithm includes: a simulated annealing algorithm, a particle swarm algorithm, and a grey wolf algorithm.

6. An intelligent reinforcement device for compression components based on big data, characterized in that: include: An image acquisition unit, used to receive a high-definition image of the surface of a concrete pier uploaded by a smart handheld terminal, wherein the compression member is a pier to be reinforced; A friction coefficient analysis unit, which is used to identify the friction coefficient between the surface of the bridge pier to be reinforced and the steel wire rope at the preset marking points by using an image recognition module based on a convolutional neural network, and generate a friction coefficient matrix, wherein the preset marking points are arranged at intervals on the steel wire rope to form a matrix position layout with m rows and n columns; The pier cracking stress analysis unit is used to input the trained LSTM neural network model based on the sunshine, temperature, humidity, overload level and overload frequency information of the pier to be reinforced, and obtain the cracking stress σ_0 of the pier to be reinforced; A reinforcement scheme analysis unit is used to calculate a reinforcement scheme based on the friction coefficient matrix of the marked points and the cracking stress of the bridge pier to be reinforced, wherein the reinforcement scheme includes the determined segment length of the steel wire rope, the vertical spacing of the steel wire rope, the target friction coefficient of the concrete surface, and the steel wire rope model, and transmit the reinforcement scheme to the smart handheld terminal; The reinforcement scheme is determined based on minW, where: ; ； ； ； The constraints are: ; ； ； in: W——Main cost of reinforcement plan, in ten thousand yuan; ——The cost of reducing the friction coefficient of concrete in the reinforcement scheme, in ten thousand yuan, The cost required to reduce the friction coefficient of the concrete surface by 1 in an area of ​​0.2m×0.1m; ——Material and operating costs of the segmented tension anchorage devices in the reinforcement scheme, in ten thousand yuan, The material and operating costs of the segmented tensioning anchorage device generated each time the 360° wire rope is split; ——Material and labor costs of prestressed steel wire ropes in the reinforcement scheme, in ten thousand yuan, The material and labor cost for each 1m of prestressed steel wire rope used in the reinforcement scheme; p——The number of points in the friction coefficient matrix where the friction coefficient needs to be reduced. Assume that the number of rows of points in the friction coefficient matrix where the friction coefficient needs to be reduced is ; - The friction coefficient of the marking point where the friction coefficient needs to be reduced; ——The friction coefficient of the concrete surface after reinforcement scheme treatment, that is, the target friction coefficient; D——diameter of the cross section of the concrete pier to be reinforced, in meters; R——radius of the cross section of the concrete pier to be reinforced, in m; H——vertical height of the area to be reinforced of the concrete pier, in meters; h——vertical spacing of steel wire ropes in the reinforcement scheme, in m; a——the number of vertically distributed steel wire ropes in the reinforcement scheme; ——The segmented steel wire rope in the reinforcement scheme corresponds to 1 / 2 of the central angle of the circle, in radians. It is the minimum limit value of the arc corresponding to the wire rope segmentation; l——the length of the steel wire rope segment in the reinforcement scheme, in meters; σ——the minimum compressive stress that can be formed on the surface of the concrete pier by the reinforcement scheme, in MPa; ——Prestressed steel wire rope tensioning control stress, unit is MPa; ——The maximum allowable stress of prestressed steel wire rope, in MPa.

7. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

8. An intelligent reinforcement system for compression components based on big data, characterized in that: include: Segmented prestressed steel wire ropes are used to reinforce compression members, which are concrete bridge piers; An intelligent segmented tensioning and anchoring module, comprising a tensioning and anchoring mechanical device and an intelligent control system, wherein the tensioning and anchoring mechanical device is connected to the segmented prestressed steel wire rope, and the intelligent control system is connected to the tensioning and anchoring mechanical device, and is used to perform bidirectional tensioning and anchoring on the segmented prestressed steel wire rope by controlling the tensioning and anchoring mechanical device; The smart handheld terminal is connected to the intelligent control system for interactively transmitting tensioning and anchoring data to obtain tensioning force and anchoring force, and monitor the entire tensioning process; it is also used to take ultra-high-pixel photos of the surface of the concrete pier; the smart handheld terminal is also connected to the cloud server for uploading data collected by the smart handheld terminal and receiving data sent by the cloud server; The cloud server is communicatively connected to the smart handheld terminal, and is used to execute any of the methods described in claims 1-5 based on the data uploaded by the smart handheld terminal to determine the reinforcement plan and send it to the smart handheld terminal, and is also used to monitor the entire tensioning process based on the data uploaded by the smart handheld terminal.

9. The intelligent reinforcement system for compression members based on big data according to claim 8 is characterized in that: The intelligent handheld terminal comprises: High-precision GPS positioning module, used to locate the concrete piers to be reinforced; The automated modeling module is used to generate the pier structure to be reinforced and the regional scene model based on the pier height, diameter, sunlight obstruction, and material property information collected on site.

Images