A curtain wall cable force detection method, device, equipment and storage medium

Abstract

The application relates to the technical field of building engineering and discloses a curtain wall cable force detection method, device, equipment and storage medium, the method comprising the following steps: constructing a finite element model of a cable-frame structure based on the structural parameters of a cable in a target curtain wall structure; generating cable shape-cable force data based on the finite element model and the structural parameters; training a model by using the cable shape-cable force data to obtain a cable force detection model; obtaining three-dimensional scanning data of the cable and the frame to generate cable point cloud and frame point cloud; and detecting based on the cable point cloud, the frame point cloud and the structural parameters by using the cable force detection model to obtain a target cable force. The finite element model of the cable-frame is constructed to accurately simulate the stress state of the cable, the cable force detection model is trained to accurately fit the nonlinear relationship between the cable shape and the cable force, the cable force can be quickly and accurately detected, the detection precision is improved, the detection cost is reduced, multiple cables can be detected at the same time, and the detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and in particular to a curtain wall cable force detection method, device, equipment and storage medium. Background Art

[0002] Cables, by applying prestress, transfer the external loads borne by the glass panels to the steel structural frames on both sides. They are one of the most widely used curtain wall support systems. During operation, cable tension can fluctuate abnormally over time, potentially causing safety incidents. Therefore, accurately measuring cable tension is essential for curtain wall health monitoring.

[0003] Existing cable tension detection methods are primarily categorized as pressure, frequency, and cable shape. The pressure method employs a contact force transmission mechanism, employing force or stress sensors to directly measure cable tension. This method primarily involves two approaches: hydraulic sensing systems and through-hole force rings. The frequency method, based on solving the string vibration differential equation, infers cable tension by obtaining the cable's natural frequency through spectral analysis. The cable shape method, a non-contact method, inverts cable tension using cable geometry parameters.

[0004] However, existing methods have significant drawbacks: The pressure method requires simultaneous installation of sensors during the construction phase, and their performance degrades significantly with long-term exposure to the environment, resulting in a short lifespan and high costs. The frequency method relies on idealized theoretical assumptions, resulting in significant errors in the bending stiffness of short cables and being susceptible to interference from environmental noise and multi-cable coupled vibrations. The cable shape method suffers from geometric nonlinear simplification errors and neglects bending stiffness, leading to insufficient detection accuracy for long cables or complex cable structures. In summary, existing methods generally suffer from low detection precision and accuracy, making them difficult to meet the engineering needs of long-term health monitoring of curtain wall cables. Summary of the Invention

[0005] In view of this, the present invention provides a curtain wall cable tension detection method, device, equipment and storage medium to solve the problems of low detection precision and poor accuracy commonly found in existing cable tension detection methods.

[0006] In a first aspect, the present invention provides a curtain wall cable force detection method, the method comprising:

[0007] Based on the structural parameters of the cables in the target curtain wall structure, a finite element model of the cable-frame structure is constructed;

[0008] Generate multiple cable shape and cable force data based on the finite element model and structural parameters;

[0009] A cable force detection model is obtained by using multiple cable shape-cable force data for model training;

[0010] Obtain 3D scanning data of the cables and frames in the target curtain wall structure, and generate cable point clouds and frame point clouds based on the 3D scanning data of the cables and frames;

[0011] The cable force detection model is used to perform detection based on the cable point cloud, frame point cloud and structural parameters to obtain the target cable force.

[0012] The present invention constructs a finite element model of the cable-frame structure to accurately simulate the stress state of the cable in the actual environment. The finite element model generates a variety of cable shapes and corresponding cable force data, which can cover the mechanical behavior of the cable under different working conditions. Using these data as a training set can improve the generalization ability of the subsequent cable force detection model. By training the cable force detection model, the model can accurately fit the nonlinear relationship between cable shape and cable force. In actual application, the three-dimensional scanning data of the cables and frames in the target curtain wall structure are obtained, and the corresponding point cloud is generated to comprehensively describe the spatial distribution of the cables and frames. The trained cable force detection model can be used to quickly and accurately predict the cable force value of the target cable, which greatly improves the accuracy of cable force detection. No expensive special equipment is required, which reduces the detection cost. In addition, cable force detection can be performed on multiple cables at the same time, thereby improving detection efficiency.

[0013] In an optional embodiment, the structural parameters include cross-sectional diameter, elastic modulus, material density, cable span, and connection method between the cable and the frame;

[0014] Based on the structural parameters of the cables in the target curtain wall structure, a finite element model of the cable-frame structure is constructed, including:

[0015] Calculate the cross-sectional area of ​​the cable based on its cross-sectional diameter;

[0016] Calculate the deadweight load per unit length of the cable based on its material density and cross-sectional area;

[0017] A cable unit model is constructed based on the cross-sectional area, elastic modulus and self-weight load per unit length of the cable;

[0018] Construct a frame unit model based on the connection between the cables and the frame and the cable span;

[0019] Based on the cable unit model and the frame unit model, a finite element model of the cable-frame structure is constructed.

[0020] The present invention calculates the cross-sectional area and self-weight load per unit length of the cable to accurately reflect the actual force-bearing capacity of the cable and the influence of the cable's own weight on the overall structure. The cable unit model is constructed in combination with the cross-sectional area, elastic modulus and self-weight load per unit length of the cable, which can comprehensively describe the mechanical properties of the cable. The frame unit model is constructed based on the connection method between the cable and the frame and the cable span, which can accurately describe the supporting effect of the frame on the cable. The cable unit model and the frame unit model are organically combined to form a complete cable-frame structure finite element model, which provides a reliable basis for the subsequent generation of cable shape-cable force data.

[0021] In an optional embodiment, the structural parameters further include sag and cable end inclination;

[0022] Based on the finite element model and structural parameters, multiple cable shape and cable force data are generated, including:

[0023] The horizontal cable force is calculated based on the cable span, sag and self-weight load per unit length using a finite element model.

[0024] Calculate the axial cable force based on the horizontal cable force and the cable end inclination angle;

[0025] The catenary equation embedded in the finite element model is used to generate the cable curve based on the horizontal cable force, self-weight load per unit length and cable span;

[0026] Keeping the cable span and self-weight load per unit length unchanged, traverse multiple sags and cable end inclinations to generate the corresponding axial cable force and cable shape curves;

[0027] For each combination of sag and cable end inclination, the corresponding axial cable force and cable shape curve are modified to obtain the target axial cable force and target cable shape curve;

[0028] The cable span, sag, cable end inclination, self-weight load per unit length, target axial cable force and target cable shape curve corresponding to each combination are taken as cable shape-cable force data.

[0029] The present invention can accurately reflect the stress state of the cable under different working conditions by calculating the horizontal cable force. Combining the horizontal cable force and the cable end inclination angle, the axial cable force of the cable can be further derived. The catenary equation embedded in the finite element model is used to generate a cable shape curve, which can intuitively reflect the shape change of the cable under different working conditions, keep the cable span and the self-weight load per unit length unchanged, traverse multiple combinations of sag and cable end inclination angles, and generate diverse axial cable forces and cable shape curves to cover the mechanical behaviors under different working conditions, thereby improving the generalization ability of the subsequent cable force detection model, and improving the reliability of the training data by correcting each combination of sag and cable end inclination angle.

[0030] In an optional embodiment, for each combination of sag and cable end inclination angle, the corresponding axial cable force and cable shape curve are modified to obtain the target axial cable force and target cable shape curve, including:

[0031] The finite element model is used to calculate the geometric stiffness matrix based on the axial cable force and cable span;

[0032] Calculate the elastic stiffness matrix based on the cable span, cross-sectional area and elastic modulus of the cable;

[0033] The geometric stiffness matrix is ​​coupled with the elastic stiffness matrix to solve the node displacement. The node displacement is the elongation of the discretized nodes constituting the finite element model along the cable axis and the sag change perpendicular to the cable direction under the action of load.

[0034] Update axial cable forces and cable shape curves based on node displacements;

[0035] The above updating process is repeated until the relative change rate of the axial cable force before and after the update is less than the preset threshold value, and the axial cable force after the last update is used as the target axial cable force, and the cable shape curve after the last update is used as the target cable shape curve.

[0036] The present invention can reflect the geometric nonlinear characteristics of the cable under different working conditions and accurately describe the material rigidity characteristics of the cable by calculating the geometric stiffness matrix and the elastic stiffness matrix. The two matrices are coupled to solve the node displacement, and the geometric nonlinearity and material characteristics of the cable are simultaneously considered, and the deformation state of the cable is fully reflected. The axial cable force and cable shape curve are updated based on the node displacement, and the actual force and geometric state of the cable can be gradually approached. The axial cable force and cable shape curve are gradually updated by an iterative method until the convergence condition is met, which significantly improves the reliability of the result.

[0037] In an optional embodiment, a plurality of cable shape-cable force data are used to perform model training to obtain a cable force detection model, including:

[0038] Divide multiple cable shape-cable force data into training sets and test sets;

[0039] Using the initial cable force detection model, the detection is performed based on any cable shape-cable force data in the training set to obtain the training cable force value;

[0040] Calculate the training error based on the difference between the target axial cable force and the training cable force value based on the cable shape-cable force data;

[0041] Based on the training error, the initial cable force detection model is optimized, and the above model training process is repeated until the preset stopping condition is reached. The initial cable force detection model obtained by the last optimization is used as the cable force detection model.

[0042] The present invention uses an initial cable force detection model to detect any cable shape-cable force data in the training set to obtain a training cable force value, and then calculates a training error based on the difference between the target axial cable force and the training cable force value. The initial cable force detection model is optimized based on the training error, which can gradually improve the prediction accuracy of the model. The above model training process is repeated until the preset stopping condition is reached, which significantly improves the convergence and stability of the model, so that the final trained model has high prediction accuracy and generalization ability, and can accurately reflect the mechanical properties of the cable.

[0043] In an optional embodiment, a cable force detection model is used to detect the target cable force based on the cable point cloud, the frame point cloud, and the structural parameters, including:

[0044] Extract the cable axis and frame axis based on the cable point cloud and frame point cloud;

[0045] Determine the end displacement of the cable based on the difference between the frame axis and the theoretical frame axis;

[0046] Extract the cable sag and cable length based on the cable axis;

[0047] The end displacement, cable sag and cable length are input into the cable force detection model to obtain the target cable force.

[0048] The present invention extracts the cable axis and the frame axis based on the cable point cloud and the frame point cloud, which can accurately reflect the actual geometric shapes of the cable and the frame, determines the end displacement of the cable based on the difference between the frame axis and the theoretical frame axis, and can capture the displacement change of the cable under actual working conditions, and extracts the cable sag and cable length based on the cable axis, which can comprehensively describe the geometric characteristics of the cable. Finally, the end displacement, cable sag and cable length are input into the cable force detection model, which can quickly and accurately obtain the target cable force of the cable.

[0049] In an optional embodiment, extracting the cable axis and the frame axis based on the cable point cloud and the frame point cloud includes:

[0050] intercepting a plurality of cross-sectional point clouds based on the cable point cloud, and determining the centroid coordinates of each cross-sectional point cloud based on a plurality of cross-sectional outer edge points of the cross-sectional point cloud;

[0051] Determine the centroid coordinates and cross-sectional area of ​​a triangular mesh consisting of the centroid coordinates and a plurality of cross-sectional outer edge points;

[0052] Determine the centroid coordinates of the cross-sectional point cloud based on the cross-sectional areas and centroid coordinates of all triangular meshes in the cross-sectional point cloud;

[0053] Fit the centroid coordinates of all cross-section point clouds to obtain the cable axis;

[0054] Based on the frame point cloud, multiple cross-sectional point clouds are intercepted. For each cross-sectional point cloud, the average value of all coordinates in the cross-sectional point cloud is used as the centroid coordinate of the cross-sectional point cloud.

[0055] The centroid coordinates of all cross-sectional point clouds are fitted to obtain the frame axis.

[0056] The present invention ensures the accurate characterization of the cross-sectional shape through the combined use of cross-sectional area and centroid coordinates. Based on the cross-sectional area and centroid coordinates of all triangular meshes, the centroid coordinates of the cross-sectional point cloud are determined. Finally, the centroid coordinates of all cross-sectional point clouds are fitted to obtain the cable axis, which can accurately describe the overall geometric shape of the cable. Based on the frame point cloud, multiple cross-sectional point clouds are intercepted, and the average value of all coordinates in each cross-sectional point cloud is used as the centroid coordinate, which can quickly and accurately extract the frame axis.

[0057] In a second aspect, the present invention provides a curtain wall cable force detection device, the device comprising:

[0058] A construction module for constructing a finite element model of a cable-frame structure based on the structural parameters of the cables in the target curtain wall structure;

[0059] A first generating module is used to generate a plurality of cable shape-cable force data based on the finite element model and structural parameters;

[0060] A training module is used to perform model training using multiple cable shape-cable force data to obtain a cable force detection model;

[0061] The second generation module is used to obtain the three-dimensional scanning data of the cables and frames in the target curtain wall structure, and generate the cable point cloud and the frame point cloud based on the three-dimensional scanning data of the cables and frames;

[0062] The detection module is used to adopt the cable force detection model to perform detection based on the cable point cloud, frame point cloud and structural parameters to obtain the target cable force.

[0063] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the curtain wall cable force detection method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0064] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the curtain wall cable force detection method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the specific embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0066] Figure 1 is a flow chart of the curtain wall cable force detection method according to an embodiment of the present application;

[0067] Figure 2 is an example diagram of a finite element model according to an embodiment of the present application;

[0068] Figure 3 is a site example diagram of a three-dimensional scanning device according to an embodiment of the present application;

[0069] Figure 4 is an example diagram of an initial cable force detection model according to an embodiment of the present application;

[0070] Figure 5 is a schematic diagram of a cable axis according to an embodiment of the present application;

[0071] Figure 6 is a schematic diagram of a cross-sectional point cloud according to an embodiment of the present application;

[0072] Figure 7 is a schematic diagram of another cable axis according to an embodiment of the present application;

[0073] Figure 8 is a structural block diagram of the curtain wall cable force detection device according to an embodiment of the present application;

[0074] Figure 9 is a hardware structure schematic diagram of the computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0075] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0076] Existing cable force detection methods generally have the problems of low detection precision and poor accuracy, which makes it difficult to meet the engineering requirements of long-term health monitoring of curtain wall cables. The present invention can accurately simulate the stress state of cables in actual environments by constructing a finite element model of the cable-frame structure. A variety of cable shapes and corresponding cable force data are generated through the finite element model. The cable force detection model is trained with these data. The model can accurately fit the nonlinear relationship between cable shape and cable force. In actual application, the three-dimensional scanning data of the cables and frames in the target curtain wall structure are obtained, and the corresponding point cloud is generated to fully describe the spatial distribution of the cables and frames. The trained cable force detection model can quickly and accurately predict the cable force value of the target cable, greatly improving the accuracy of cable force detection. It does not require expensive special equipment, reduces detection costs, and can simultaneously perform cable force detection on multiple cables, improving detection efficiency.

[0077] According to an embodiment of the present invention, an embodiment of a curtain wall cable tension detection method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0078] In this embodiment, a curtain wall cable force detection method is provided. Figure 1 FIG. 1 is a flow chart of a curtain wall cable force detection method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0079] Step S101 constructs a finite element model of the cable-frame structure based on the structural parameters of the cables in the target curtain wall structure. Specifically, parameters such as the cable cross-sectional diameter, elastic modulus, material density, and cable span are obtained from design drawings or on-site surveys. The connection method between the cables and the frame (e.g., fixed supports or sliding hinge supports) is also determined. This data provides the foundation for finite element modeling, ensuring that the model accurately reflects the physical properties and boundary conditions of the cables. Figure 2 is an example diagram of a finite element model according to an embodiment of the present invention, such as Figure 2 As shown, the finite element model includes cable components and a frame component, which is used to simulate the stress conditions of the curtain wall structure under different working conditions. By establishing a calculation model that accurately reflects the characteristics of the curtain wall structure, it lays the foundation for the subsequent generation of cable shape and cable force data.

[0080] In step S102, a plurality of cable shape-cable force data are generated based on the finite element model and the structural parameters. Specifically, a plurality of related formulas for calculating cable force are embedded in the finite element model. Based on the related factors affecting the cable force, the formulas can be used to construct complete cable shape-cable force data. Meanwhile, by changing these factors affecting the cable force, the changes of the cable shape-cable force under different working conditions can be simulated, providing rich samples for model training, so that the trained model can adapt to various actual situations and improve the generalization ability of the model.

[0081] In step S103, the model is trained using a plurality of cable shape-cable force data to obtain a cable force detection model. Specifically, the pressure method, frequency method and cable shape method used in related technologies cannot accurately measure the cable force. In the embodiment of the present application, the deep learning neural network is trained using cable shape-cable force data, so that the model can accurately fit the nonlinear relationship between the cable shape and the cable force, thereby improving the precision and efficiency of cable force detection.

[0082] In step S104, three-dimensional scanning data of the cable and the frame in the target curtain wall structure are obtained, and based on the three-dimensional scanning data of the cable and the frame, cable point cloud and frame point cloud are generated. Specifically, in actual application, three-dimensional scanning equipment such as a three-dimensional scanner is used to collect three-dimensional scanning data. Figure 3 is a site example diagram of the three-dimensional scanning equipment according to the embodiment of the present application, as Figure 3 shown, the three-dimensional scanning equipment is located at a plurality of sites, each site needs to cover the complete field of view of the cable and the frame, the adjacent sites have an overlap rate of ≥30%, the three-dimensional scanning data of the cable and the frame are collected from multiple angles, and based on the three-dimensional scanning data, the cable point cloud and the frame point cloud are generated through point cloud registration, noise reduction filtering and point cloud classification, so as to accurately reflect the actual shape of the cable and the frame and provide reliable data for subsequent cable force detection. Optionally, the above point cloud generation process can be realized by using existing technologies, and thus will not be described here.

[0083] In step S105, the cable force detection model is used to detect based on the cable point cloud, the frame point cloud and the structural parameters to obtain the target cable force. Specifically, the trained cable force detection model can accurately detect the cable force according to the cable point cloud and the frame point cloud, thereby improving the precision and efficiency of cable force detection, and without the participation of sensors and other devices, the detection cost is effectively reduced.

[0084] The present invention constructs a finite element model of the cable-frame structure to accurately simulate the stress state of the cable in the actual environment. The finite element model generates a variety of cable shapes and corresponding cable force data, which can cover the mechanical behavior of the cable under different working conditions. Using these data as a training set can improve the generalization ability of the subsequent cable force detection model. By training the cable force detection model, the model can accurately fit the nonlinear relationship between cable shape and cable force. In actual application, the three-dimensional scanning data of the cables and frames in the target curtain wall structure are obtained, and the corresponding point cloud is generated to comprehensively describe the spatial distribution of the cables and frames. The trained cable force detection model can be used to quickly and accurately predict the cable force value of the target cable, which greatly improves the accuracy of cable force detection. No expensive special equipment is required, which reduces the detection cost. In addition, cable force detection can be performed on multiple cables at the same time, thereby improving detection efficiency.

[0085] In this embodiment, a curtain wall cable tension detection method is provided, which specifically includes the following steps:

[0086] Step S401: constructing a finite element model of a cable-frame structure based on the structural parameters of the cables in the target curtain wall structure. The structural parameters include cross-sectional diameter, elastic modulus, material density, cable span, and the connection method between the cables and the frame.

[0087] Specifically, the above step S401 includes:

[0088] In step S4011, the cross-sectional area of ​​the cable is calculated based on the cross-sectional diameter of the cable. Specifically, the cross-sectional area directly affects the axial stiffness and deadweight load of the cable, and can be calculated using the following formula (1).

[0089] (1)

[0090] in, represents the cross-sectional area; Indicates the cross-sectional diameter.

[0091] In step S4012, the cable's self-weight load per unit length is calculated based on the cable's material density and cross-sectional area. Specifically, determining the cable's self-weight load distribution is a key parameter for calculating the cable's shape and can be calculated using the following formula (2).

[0092] (2)

[0093] in, Indicates the self-weight load per unit length; Indicates the material density; represents the cross-sectional area; Represents the acceleration due to gravity.

[0094] In step S4013, a cable element model is constructed based on the cable's cross-sectional area, elastic modulus, and deadweight load per unit length. Specifically, the cable element in finite element software is used. By inputting the cross-sectional area, elastic modulus, and deadweight load per unit length, a mechanical calculation model of the cable is established to simulate its tensile properties and geometric nonlinear behavior.

[0095] Step S4014: Based on the cable-to-frame connection method and cable span, a frame unit model is constructed. Specifically, using the frame unit in finite element software, the cable span and connection method are input to establish a mechanical model of the supporting frame, simulating its boundary constraints on the cables and load transfer.

[0096] In step S4015, a finite element model of the cable-frame structure is constructed based on the cable unit model and the frame unit model. Specifically, the cable unit endpoints are bound to the frame unit nodes through node coupling or rigid connection to ensure load transfer continuity. This integrates the cable units and the frame units to form a complete curtain wall structure calculation model.

[0097] Step S402: Generate a plurality of cable shape-cable force data based on the finite element model and structural parameters. The structural parameters also include sag and cable end inclination.

[0098] Specifically, the above step S402 includes:

[0099] In step S4021, the horizontal cable force is calculated using a finite element model based on the cable span, sag, and deadweight load per unit length. Specifically, using the catenary theory formula embedded in the finite element model, when the cable is subjected to deadweight load per unit length, the relationship between the horizontal cable force, sag, and cable span can be expressed as follows: (3) Therefore, the horizontal cable force can be calculated using (3) to reflect the cable's ability to resist deadweight sag, which is a key factor in cable force analysis.

[0100] (3)

[0101] in, represents the horizontal cable force; Indicates the self-weight load per unit length; Indicates rope span; Indicates sag.

[0102] In step S4022, the axial cable force is calculated based on the horizontal cable force and the cable end inclination angle. Specifically, the axial cable force, which represents the actual tension on the cable, can be calculated using the following equation (4). Converting the horizontal cable force into the total axial tension of the cable, and considering the effect of the cable end inclination angle on the tension, more intuitively reflects the actual stress state of the cable.

[0103] (4)

[0104] in, represents the axial cable force; represents the horizontal cable force; Indicates the inclination angle of the cable end.

[0105] In step S4023, the catenary equation embedded in the finite element model is used to generate a cable curve based on the horizontal cable force, the deadweight load per unit length, and the cable span. Specifically, the catenary equation embedded in the finite element model is used, as shown in Equation (5) below, to calculate the vertical displacement corresponding to the coordinates of each point on the cable, thereby generating a smooth cable curve that intuitively displays the drooping shape of the cable.

[0106] (5)

[0107] in, Indicates the horizontal coordinate; represents the vertical displacement; Indicates the self-weight load per unit length; Represents the horizontal cable force.

[0108] Step S4024: Maintaining the cable span and deadweight load per unit length, multiple sags and cable end inclinations are traversed to generate corresponding axial cable force and shape curves. Specifically, the cable span and deadweight load per unit length are fixed, and parameter combinations are traversed with a preset step size (e.g., sag from 0.1L to 0.3L, step size 0.05L; cable end inclination from 5° to 30°, step size 5°). Steps S4021-S4023 are repeated for each combination to calculate the corresponding horizontal cable force, axial cable force, and cable shape curves. This allows simulation of cable force and shape variations under different operating conditions, obtaining diverse training samples, and improving model generalization.

[0109] Step S4025: For each combination of sag and cable end inclination angle, the corresponding axial cable force and cable shape curve are corrected to obtain the target axial cable force and target cable shape curve.

[0110] In some optional implementations, the above step S4025 includes:

[0111] In step S40251, a finite element model is used to calculate the geometric stiffness matrix based on the axial cable force and cable span. Specifically, considering the effect of cable tension on structural stiffness, the geometric stiffness matrix is ​​used to describe the stiffness change caused by the axial force. It reflects the ability of the axial tension to resist bending in the cable. The greater the tension, the higher the stiffness. Optionally, the geometric stiffness matrix can be expressed as the following equation (6).

[0112] (6)

[0113] in, represents the geometric stiffness matrix; represents the axial cable force; denotes the cable span.

[0114] At step S40252, the elastic stiffness matrix is calculated based on the cable span, the cross-sectional area and the elastic modulus of the cable. Specifically, the elastic stiffness matrix can be represented by the following formula (7), which reflects the elastic elongation characteristics of the cable under axial load.

[0115] (7)

[0116] wherein, denotes the elastic stiffness matrix; denotes the elastic modulus; denotes the cross-sectional area; denotes the cable span.

[0117] At step S40253, the geometric stiffness matrix and the elastic stiffness matrix are coupled to solve the node displacement, which is the elongation of the discretized node constituting the finite element model along the axial direction of the cable and the sag change perpendicular to the cable direction under the action of the load. Specifically, the above two matrices are coupled to obtain an overall stiffness matrix, so as to solve the node displacement based on the finite element equation shown in the following formula (8). Wherein, the node displacement includes the axial elongation and the sag change. By coupling the geometric stiffness and the elastic stiffness, the model can simulate the tension-displacement nonlinear relationship under large deformation, so as to solve the real deformation of the cable under the action of the load.

[0118] (8)

[0119] wherein, denotes the overall stiffness matrix; denotes the node displacement; denotes the load vector.

[0120] At step S40254, the axial cable force and the cable shape curve are updated based on the node displacement. Specifically, the cable elongation is calculated from the node displacement, and the axial cable force is corrected by the Hooke's law shown in the following formula (9). The catenary equation parameters are adjusted according to the sag change, and the cable shape curve is regenerated. By updating the cable tension and the shape according to the node displacement, it is ensured that the data conforms to the actual mechanical response, and the accuracy of the training data is improved.

[0121] (9)

[0122] wherein, denotes the axial cable force; denotes the elastic modulus; denotes the cross-sectional area; denotes the cable elongation; denotes the cable span.

[0123] Step S40255 repeats the above updating process until the relative rate of change of the axial cable force before and after the update is less than a preset threshold. The last updated axial cable force is used as the target axial cable force, and the last updated cable shape curve is used as the target cable shape curve. Specifically, after each axial cable force update, the relative rate of change before and after the update is calculated. If this relative rate of change is less than the preset threshold, the revised axial cable force and cable shape curve meet the accuracy requirements. The axial cable force and cable shape curve obtained from the last update are used as the final target axial cable force and cable shape curve. Through iterative correction, errors caused by the small sag assumption and linear analysis are eliminated, making the data more accurate to the actual project conditions.

[0124] In step S4026, the cable span, sag, cable end inclination, deadweight load per unit length, target axial cable force, and target cable shape curve corresponding to each combination are used as cable shape-force data. Specifically, the calculation results for each parameter combination are organized into structured data to provide standardized samples for subsequent deep learning model training.

[0125] Step S403: using a plurality of cable shape-cable force data to perform model training to obtain a cable force detection model.

[0126] Specifically, the above step S403 includes:

[0127] Step S4031: Divide the cable shape-cable force data into a training set and a test set. Specifically, the cable shape-cable force data are divided into two parts according to a preset ratio. The training set is used for model parameter optimization, and the test set is used to evaluate the model generalization ability to avoid overfitting.

[0128] Step S4032: Using the initial cable force detection model, perform detection based on any cable shape-cable force data in the training set to obtain a training cable force value. Specifically, Figure 4 An example diagram of an initial cable force detection model according to an embodiment of the present invention is shown in FIG. Figure 4 As shown in the figure, the initial cable force detection model adopts a four-layer neural network architecture (input layer-double hidden layer-output layer). Figure 4 This is a general four-layer neural network architecture. The number of input and output layer nodes is only an example. In actual applications, this can be adaptively adjusted based on the input and output data dimensions, and this is not a limitation in the present embodiment. In this embodiment, the input layer has five nodes (cable span, sag, cable end inclination, deadweight load per unit length, and target cable shape curve), and the output layer has one node (training cable force value).

[0129] At step S4033, a training error is calculated based on a difference between the target axial cable force and the training cable force value based on the cable shape-cable force data. Specifically, taking the mean square error as an error indicator, a difference between the actual target axial cable force of the cable and the training cable force value predicted by the model is calculated to obtain the training error.

[0130] At step S4034, the initial cable force detection model is optimized based on the training error, and the above model training process is repeated until a preset stopping condition is reached, and the initial cable force detection model obtained by the last optimization is taken as the cable force detection model. Specifically, the neural network weights and biases are adjusted by the gradient descent method to minimize the training error. If the mean square error decreases by less than a preset threshold in a preset number of consecutive iterations, it is considered that the model performance meets the requirements, and the model parameters (weight matrix, bias vector) optimized last time are saved to form the final cable force detection model. The nonlinear relationship between the cable shape and the cable force is fitted by deep learning, so that the trained model can accurately and quickly process the measured cable shape data.

[0131] In some optional embodiments, after obtaining the cable force detection model, the performance of the model can be verified by a verification set, which will not be described here.

[0132] At step S404, three-dimensional scanning data of the cable and the frame in the target curtain wall structure are obtained, and cable point clouds and frame point clouds are generated based on the three-dimensional scanning data of the cable and the frame. For details, please refer to Figure 1 The step S404 of the embodiment shown will not be described here.

[0133] At step S405, the cable force detection model is used to detect based on the cable point clouds, the frame point clouds and the structure parameters to obtain the target cable force.

[0134] Specifically, the above step S405 includes:

[0135] At step S4051, the cable axis and the frame axis are extracted based on the cable point clouds and the frame point clouds.

[0136] In some optional embodiments, the above step S4051 includes:

[0137] At step S40511, a plurality of cross-sectional point clouds are intercepted based on the cable point clouds, and for each cross-sectional point cloud, the centroid coordinates of the cross-sectional point cloud are determined based on a plurality of cross-sectional outer edge points of the cross-sectional point cloud. Specifically, a plurality of cross-sectional point clouds are obtained by intercepting cross sections at fixed intervals along the length direction of the cable, each cross-sectional point cloud has a plurality of cross-sectional outer edge points, and for each cross-sectional point cloud, the centroid coordinates are calculated by the following formula (10).

[0138] (10)

[0139] wherein, represents the coordinates of the center of mass; Indicates the number of outer edge points of the cross section; Indicates the The coordinates of the outer edge points of the cross section.

[0140] Step S40512: Determine the centroid coordinates and cross-sectional area of ​​a triangular mesh formed by the centroid coordinates and multiple cross-sectional outer edge points. Specifically, determine two cross-sectional outer edge points adjacent to the centroid coordinates. These three points can form a triangular mesh, and calculate the centroid coordinates of the triangular mesh using the following equation (11).

[0141] (11)

[0142] in, Represents the centroid coordinates of the triangular mesh; represents the coordinates of the center of mass; and Represents the coordinates of the two outer edge points of the cross section adjacent to the centroid coordinates.

[0143] The cross-sectional area of ​​the triangular mesh is calculated using the coordinates of the above three points using the following formula (12).

[0144] (12)

[0145] in, Represents the cross-sectional area of ​​the triangular mesh; , , , , , .

[0146] In step S40513, the centroid coordinates of the cross-sectional point cloud are determined based on the cross-sectional areas and centroid coordinates of all triangular meshes in the cross-sectional point cloud. Specifically, the centroid coordinates of the cross-sectional point cloud are determined based on the centroid coordinates and cross-sectional area of ​​each triangular mesh using the following equation (13).

[0147] (13)

[0148] in, Represents the centroid coordinates of the cross-sectional point cloud; Indicates the The centroid coordinates of the triangle mesh; Indicates the The cross-sectional area of ​​a triangle mesh; Indicates the number of triangle meshes in the cross-section point cloud.

[0149] Step S40514: Fit the centroid coordinates of all cross-sectional point clouds to obtain the cable axis. Specifically, the centroid coordinates of all cross-sectional point clouds are fitted into a spatial curve using the least squares method to obtain the cable axis.

[0150] In some optional embodiments, Figure 5 A schematic diagram of the cable axis according to an embodiment of the present invention is shown in FIG. Figure 5 As shown, the left side shows the actual shape of the cable, and the right side shows the cable axis obtained by fitting the centroid coordinates of each cross-section point cloud of the cable ( Figure 5 (red line in the figure).

[0151] In some optional embodiments, Figure 6 A schematic diagram of a cross-sectional point cloud according to an embodiment of the present invention is shown in FIG. Figure 6 As shown in the figure, the green points on the outermost layer of the cross-section point cloud are the outer edge points of the cross-section. The centroid coordinates of the cross-section point cloud can be calculated through these outer edge points of the cross-section ( Figure 6 The blue point in the figure). Through the centroid coordinates and the two adjacent cross-section outer edge points, the cross-section point cloud can be divided into 12 triangular meshes, and then the centroid coordinates of each triangular mesh can be calculated ( Figure 6 Finally, based on the centroid coordinates and cross-sectional areas of all triangle meshes in the cross-sectional point cloud, the centroid coordinates of the cross-sectional point cloud are calculated ( Figure 6 red dot in the image).

[0152] In step S40515, multiple cross-section point clouds are extracted from the frame point cloud. For each cross-section point cloud, the average value of all coordinates in the cross-section point cloud is used as the centroid coordinates of the cross-section point cloud. Specifically, cross-sections are extracted at regular intervals along the length of the frame, and the centroid of each cross-section point cloud is calculated. The average value of all coordinates in each cross-section point cloud is used as the centroid coordinates of the cross-section point cloud.

[0153] Step S40516: Fit the centroid coordinates of all cross-sectional point clouds to obtain the frame axis. Specifically, referring to step S40514, the centroid coordinates of all cross-sectional point clouds are fitted into a space curve using the least squares method to obtain the frame axis.

[0154] Step S4052: Determine the end displacement of the cable based on the difference between the frame axis and the theoretical frame axis. Specifically, the end displacement reflects the actual deformation of the curtain wall structure. Extract the theoretical coordinate data of the frame from the design drawings, or generate the frame axis under ideal conditions through the finite element model. Compare the measured frame axis with the theoretical axis. Assume that the coordinates of the cable anchor point on the theoretical frame axis are , the coordinates on the frame axis are , then the end displacement is The end displacement is decomposed into the displacement component along the cable axis (affecting the cable length) and the component perpendicular to the cable direction (affecting the sag).

[0155] Step S4053 extracts the cable sag and cable length based on the cable axis. Specifically, the two endpoints of the cable axis are determined, and the horizontal projection span of the line connecting the two endpoints is calculated. The coordinate point on the cable axis that is farthest from the line is found, and the vertical distance between this farthest coordinate point and the line is calculated to obtain the cable sag. The cable axis is segmented and fitted with a straight line. The length of each segment is accumulated to obtain the cable length.

[0156] In some optional embodiments, Figure 7 A schematic diagram of another cable axis according to an embodiment of the present invention is shown in FIG. Figure 7 As shown, according to Figure 5 The cable axis shown can be used to calculate the cable sag information along the cable length, that is, the cable sag and the cable length.

[0157] In step S4054, the end displacement, cable sag, and cable length are input into the cable force detection model to obtain the target cable force. Specifically, the measured geometric parameters are input into the trained cable force detection model, and the actual cable force is predicted using a deep learning algorithm to obtain the target cable force, achieving accurate measurement of the cable force.

[0158] The present invention constructs a finite element model of the cable-frame structure to accurately simulate the stress state of the cable in the actual environment. The finite element model generates a variety of cable shapes and corresponding cable force data, which can cover the mechanical behavior of the cable under different working conditions. Using these data as a training set can improve the generalization ability of the subsequent cable force detection model. By training the cable force detection model, the model can accurately fit the nonlinear relationship between cable shape and cable force. In actual application, the three-dimensional scanning data of the cables and frames in the target curtain wall structure are obtained, and the corresponding point cloud is generated to comprehensively describe the spatial distribution of the cables and frames. The trained cable force detection model can be used to quickly and accurately predict the cable force value of the target cable, which greatly improves the accuracy of cable force detection. No expensive special equipment is required, which reduces the detection cost. In addition, cable force detection can be performed on multiple cables at the same time, thereby improving detection efficiency.

[0159] This embodiment also provides a curtain wall cable tension detection device, which is used to implement the above-mentioned embodiments and preferred implementations. Details already described will not be repeated here. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented using software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0160] This embodiment provides a curtain wall cable force detection device, such as Figure 8 Shown, including:

[0161] The construction module 801 is used to construct a finite element model of the cable-frame structure based on the structural parameters of the cables in the target curtain wall structure.

[0162] The first generating module 802 is used to generate a plurality of cable shape-cable force data based on the finite element model and structural parameters.

[0163] The training module 803 is used to perform model training using a plurality of cable shape-cable force data to obtain a cable force detection model.

[0164] The second generating module 804 is used to obtain the 3D scanning data of the cables and frames in the target curtain wall structure, and generate the cable point cloud and the frame point cloud based on the 3D scanning data of the cables and frames.

[0165] The detection module 805 is used to use a cable force detection model to perform detection based on the cable point cloud, the frame point cloud and the structural parameters to obtain the target cable force.

[0166] In some optional embodiments, the structural parameters include cross-sectional diameter, elastic modulus, material density, cable span, and connection method between the cable and the frame;

[0167] Building block 801 includes:

[0168] The first calculation unit is used to calculate the cross-sectional area of ​​the cable based on the cross-sectional diameter of the cable.

[0169] The second calculation unit is used to calculate the self-weight load per unit length of the cable based on the material density and cross-sectional area of ​​the cable.

[0170] The first construction unit is used to construct a cable unit model based on the cross-sectional area, elastic modulus and self-weight load per unit length of the cable.

[0171] The second construction unit is used to construct a frame unit model based on the connection method between the cables and the frame and the cable span.

[0172] The third construction unit is used to construct a finite element model of the cable-frame structure based on the cable unit model and the frame unit model.

[0173] In some optional embodiments, the structural parameters further include sag and cable end inclination;

[0174] The first generation module 802 includes:

[0175] The third calculation unit is used to calculate the horizontal cable force based on the cable span, sag and self-weight load per unit length of the cable using a finite element model.

[0176] The fourth calculation unit is used to calculate the axial cable force based on the horizontal cable force and the cable end inclination angle.

[0177] The first generation unit is used to generate a cable curve based on the horizontal cable force, the self-weight load per unit length, and the cable span using the catenary equation embedded in the finite element model.

[0178] The second generating unit is used to keep the cable span and the deadweight load per unit length unchanged, traverse multiple sags and cable end inclinations, and generate corresponding axial cable forces and cable shape curves.

[0179] The correction unit is used for correcting the corresponding axial cable force and cable shape curve for each combination of sag and cable end inclination angle to obtain the target axial cable force and target cable shape curve.

[0180] The first determining unit is used to use the cable span, sag, cable end inclination, unit length deadweight load, target axial cable force and target cable shape curve corresponding to each combination as cable shape-cable force data.

[0181] In some optional embodiments, the correction unit includes:

[0182] The first calculation subunit is used to calculate the geometric stiffness matrix based on the axial cable force and the cable span by using a finite element model.

[0183] The second calculation subunit is used to calculate the elastic stiffness matrix based on the cable span, cross-sectional area and elastic modulus of the cable.

[0184] The coupling subunit is used to couple the geometric stiffness matrix with the elastic stiffness matrix to solve the node displacement. The node displacement is the elongation of the discretized nodes constituting the finite element model along the cable axis and the sag change perpendicular to the cable direction under the action of load.

[0185] The first updating subunit is used to update the axial cable force and cable shape curve based on the node displacement.

[0186] The second updating subunit is used to repeat the above updating process until the relative change rate of the axial cable force before and after the update is less than a preset threshold, and the axial cable force after the last update is used as the target axial cable force, and the cable curve after the last update is used as the target cable curve.

[0187] In some optional implementations, the training module 803 includes:

[0188] The partitioning unit is used to divide multiple cable shape-cable force data into a training set and a test set.

[0189] The first detection unit is used to use an initial cable force detection model to perform detection based on any cable shape-cable force data in a training set to obtain a training cable force value.

[0190] The fifth calculation unit is used to calculate the training error based on the difference between the target axial cable force and the training cable force value of the cable shape-cable force data.

[0191] The training unit is used to optimize the initial cable force detection model based on the training error, repeat the above model training process until the preset stopping condition is reached, and use the initial cable force detection model obtained by the last optimization as the cable force detection model.

[0192] In some optional implementations, the detection module 805 includes:

[0193] The first extraction unit is used to extract the cable axis and the frame axis based on the cable point cloud and the frame point cloud.

[0194] The second determining unit is configured to determine an end displacement of the cable based on a difference between the frame axis and a theoretical frame axis.

[0195] The second extraction unit is used to extract the cable sag and the cable length of the cable based on the cable axis.

[0196] The second detection unit is used to input the end displacement, cable sag and cable length into the cable force detection model to obtain the target cable force of the cable.

[0197] In some optional embodiments, the first extraction unit includes:

[0198] The first determining subunit is configured to intercept a plurality of cross-sectional point clouds based on the cable point cloud, and determine the centroid coordinates of each cross-sectional point cloud based on a plurality of cross-sectional outer edge points of the cross-sectional point cloud.

[0199] The second determining subunit is used to determine the centroid coordinates and cross-sectional area of ​​a triangular mesh composed of the centroid coordinates and a plurality of cross-sectional outer edge points.

[0200] The third determining subunit is configured to determine the centroid coordinates of the cross-sectional point cloud based on the cross-sectional areas and centroid coordinates of all triangular meshes in the cross-sectional point cloud.

[0201] The first fitting subunit is used to fit the centroid coordinates of all cross-sectional point clouds to obtain the cable axis.

[0202] The fourth determining subunit is configured to intercept a plurality of cross-sectional point clouds based on the framework point cloud, and for each cross-sectional point cloud, take an average value of all coordinates in the cross-sectional point cloud as the centroid coordinates of the cross-sectional point cloud.

[0203] The second fitting subunit is used to fit the centroid coordinates of all cross-section point clouds to obtain the frame axis.

[0204] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0205] The curtain wall cable tension detection device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0206] The embodiment of the present invention also provides a computer device having the above Figure 8 The curtain wall cable tension detection device shown.

[0207] See also Figure 9 , Figure 9 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 9 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 9 A processor 10 is taken as an example.

[0208] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0209] The memory 20 stores instructions that can be executed by at least one processor 10, so as to enable at least one processor 10 to execute the method shown in the above embodiment.

[0210] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0211] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0212] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 can be connected via a bus or other means. Figure 9 The bus connection is taken as an example.

[0213] The input device 30 can receive input digital or character information and generate key signal input related to user settings and function control of the computer device. Examples include a touch screen, keypad, mouse, trackpad, touchpad, pointing stick, one or more mouse buttons, trackball, joystick, etc. The output device 40 may include a display device, auxiliary lighting devices (e.g., LEDs), and tactile feedback devices (e.g., vibration motors). Such display devices include, but are not limited to, liquid crystal displays, light emitting diodes, monitors, and plasma displays. In some optional embodiments, the display device may be a touch screen.

[0214] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0215] A portion of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0216] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A curtain wall cable force detection method, characterized in that: The method comprises: Constructing a finite element model of the cable-frame structure based on the structural parameters of the cables in the target curtain wall structure, wherein the structural parameters include cross-sectional diameter, elastic modulus, material density, cable span, sag, cable end inclination, and the connection method between the cables and the frame; generating a plurality of cable shape-cable force data based on the finite element model and the structural parameters; Using the plurality of cable shape-cable force data to perform model training to obtain a cable force detection model; Acquire three-dimensional scanning data of cables and frames in the target curtain wall structure, and generate cable point clouds and frame point clouds based on the three-dimensional scanning data of the cables and frames; Using the cable force detection model, detection is performed based on the cable point cloud, the frame point cloud and the structural parameters to obtain a target cable force; The step of constructing a finite element model of a cable-frame structure based on the structural parameters of the cables in the target curtain wall structure includes: Calculating the cross-sectional area of ​​the cable based on the cross-sectional diameter of the cable; Calculating the deadweight load per unit length of the cable based on the material density and cross-sectional area of ​​the cable; Constructing a cable unit model based on the cross-sectional area, elastic modulus and self-weight load per unit length of the cable; constructing a frame unit model based on the connection mode between the cables and the frame and the cable span; Constructing a finite element model of a cable-frame structure based on the cable unit model and the frame unit model; The generating of a plurality of cable shape-cable force data based on the finite element model and the structural parameters includes: Calculating the horizontal cable force based on the cable span, sag, and deadweight load per unit length of the cable using the finite element model; Calculating the axial cable force based on the horizontal cable force and the cable end inclination angle; generating a cable curve based on the horizontal cable force, the self-weight load per unit length, and the cable span using the catenary equation embedded in the finite element model; Keeping the cable span and the deadweight load per unit length unchanged, traversing multiple sags and cable end inclinations to generate corresponding axial cable forces and cable shape curves; For each combination of sag and cable end inclination, the corresponding axial cable force and cable shape curve are modified to obtain the target axial cable force and target cable shape curve; The cable span, sag, cable end inclination, self-weight load per unit length, target axial cable force and target cable shape curve corresponding to each combination are taken as cable shape-cable force data.

2. The method according to claim 1, characterized in that For each combination of sag and cable end inclination angle, the corresponding axial cable force and cable shape curve are corrected to obtain the target axial cable force and target cable shape curve, including: Using the finite element model, a geometric stiffness matrix is ​​calculated based on the axial cable force and the cable span; Calculating an elastic stiffness matrix based on the cable span, cross-sectional area, and elastic modulus of the cable; The geometric stiffness matrix is ​​coupled with the elastic stiffness matrix to solve for the node displacement, where the node displacement is the elongation of the discretized nodes constituting the finite element model along the axial direction of the cable and the sag change perpendicular to the cable direction under the action of the load; updating the axial cable force and the cable shape curve based on the node displacement; The above updating process is repeated until the relative change rate of the axial cable force before and after the update is less than a preset threshold, and the axial cable force after the last update is used as the target axial cable force, and the cable shape curve after the last update is used as the target cable shape curve.

3. The method according to claim 1, characterized in that The method of using the plurality of cable shape-cable force data to perform model training to obtain a cable force detection model includes: Dividing the plurality of cable shape-cable force data into a training set and a test set; Using an initial cable force detection model, performing detection based on any cable shape-cable force data in the training set to obtain a training cable force value; calculating a training error based on a difference between the target axial cable force of the cable shape-cable force data and the training cable force value; Based on the training error, the initial cable force detection model is optimized, and the above model training process is repeated until a preset stopping condition is reached, and the initial cable force detection model obtained by the last optimization is used as the cable force detection model.

4. The method according to claim 1, wherein The method of using the cable force detection model to detect based on the cable point cloud, the frame point cloud, and the structural parameters to obtain the target cable force includes: Extracting the cable axis and the frame axis based on the cable point cloud and the frame point cloud; determining an end displacement of the cable based on a difference between the frame axis and a theoretical frame axis; Extracting the cable sag and cable length of the cable based on the cable axis; The end displacement, the cable sag and the cable length are input into the cable force detection model to obtain the target cable force of the cable.

5. The method according to claim 4, characterized in that The extracting the cable axis and the frame axis based on the cable point cloud and the frame point cloud includes: intercepting a plurality of cross-sectional point clouds based on the cable point cloud, and determining the centroid coordinates of each cross-sectional point cloud based on a plurality of cross-sectional outer edge points of the cross-sectional point cloud; Determining the centroid coordinates and cross-sectional area of ​​a triangular mesh formed by the centroid coordinates and the plurality of cross-sectional outer edge points; Determining the centroid coordinates of the cross-sectional point cloud based on the cross-sectional areas and centroid coordinates of all triangular meshes in the cross-sectional point cloud; Fitting the centroid coordinates of all cross-sectional point clouds to obtain the cable axis; intercepting a plurality of cross-sectional point clouds based on the frame point cloud, and for each cross-sectional point cloud, taking the average value of all coordinates in the cross-sectional point cloud as the centroid coordinates of the cross-sectional point cloud; The centroid coordinates of all cross-sectional point clouds are fitted to obtain the frame axis.

6. A curtain wall cable force detection device, characterized in that: The device comprises: A construction module is used to construct a finite element model of the cable-frame structure based on the structural parameters of the cables in the target curtain wall structure, wherein the structural parameters include cross-sectional diameter, elastic modulus, material density, cable span, sag, cable end inclination angle, and connection mode between the cables and the frame; A first generating module is used to generate a plurality of cable shape-cable force data based on the finite element model and the structural parameters; A training module, configured to perform model training using the plurality of cable shape-cable force data to obtain a cable force detection model; a second generating module, configured to obtain three-dimensional scanning data of cables and frames in the target curtain wall structure, and generate a cable point cloud and a frame point cloud based on the three-dimensional scanning data of the cables and the frame; a detection module, configured to use the cable force detection model to perform detection based on the cable point cloud, the frame point cloud, and the structural parameters to obtain a target cable force; The building blocks are specifically used for: Calculating the cross-sectional area of ​​the cable based on the cross-sectional diameter of the cable; Calculating the deadweight load per unit length of the cable based on the material density and cross-sectional area of ​​the cable; Constructing a cable unit model based on the cross-sectional area, elastic modulus and self-weight load per unit length of the cable; constructing a frame unit model based on the connection mode between the cables and the frame and the cable span; Constructing a finite element model of a cable-frame structure based on the cable unit model and the frame unit model; The first generating module is specifically configured to: Calculating the horizontal cable force based on the cable span, sag, and deadweight load per unit length of the cable using the finite element model; Calculating the axial cable force based on the horizontal cable force and the cable end inclination angle; generating a cable curve based on the horizontal cable force, the self-weight load per unit length, and the cable span using the catenary equation embedded in the finite element model; Keeping the cable span and the deadweight load per unit length unchanged, traversing multiple sags and cable end inclinations to generate corresponding axial cable forces and cable shape curves; For each combination of sag and cable end inclination, the corresponding axial cable force and cable shape curve are modified to obtain the target axial cable force and target cable shape curve; The cable span, sag, cable end inclination, self-weight load per unit length, target axial cable force and target cable shape curve corresponding to each combination are taken as cable shape-cable force data.

7. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the curtain wall cable force detection method according to any one of claims 1 to 5 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the curtain wall cable force detection method according to any one of claims 1 to 5.

Images