Hexagonal-section large-cut-angle and dip-angle combined tower steel-concrete combined section construction method

Abstract

The invention relates to the field of bridge construction, and discloses a hexagonal-section large-cut-angle and dip-angle combined tower steel-concrete combined section construction method which comprises the following steps: arranging a strain sensor group and a laser calibration target spot on the outer wall of a steel shell, establishing a pose parameter mapping model through multi-source data fusion, and resolving the six-degree-of-freedom pose of the steel shell in real time; constructing a load matrix based on the strain data, and solving a target stress value of the support rod by adopting a maximum flow-minimum cut algorithm; by adjusting the telescopic amount of the telescopic supporting rod in stages and combining servo motor driving and a closed-loop feedback correction mechanism, balanced distribution of supporting force is achieved; the weight coefficient of the multi-objective optimization algorithm is dynamically adjusted, and the pose precision and the structure safety requirement are balanced; and finally, after the pose convergence and stress balance conditions are met, the driving device is dismantled, and the supporting rod body is reserved to serve as a permanent reinforcing structure. The technical problem that in a traditional method, the steel shell pose precision and the stress balance of a supporting system are difficult to consider at the same time is solved.

Description

Technical Field

[0001] The invention relates to the technical field of bridge construction, in particular to a construction method for a steel-concrete combined section of a hexagonal cross-section large-cut-angle and inclination-angle combined tower. Background Art

[0002] In recent years, steel-concrete hybrid composite bridge towers have become increasingly popular in the construction of long-span cable-stayed bridges due to their lightweight structure, high construction efficiency, and clean appearance. The steel-concrete composite section of such bridge towers, serving as the critical transition zone between the steel shell and concrete segments, must simultaneously meet stringent requirements such as the coordinated loading of dissimilar materials, high-precision matching of geometric lines, and gradual stiffness changes in complex sections. This is particularly true for the large-angle diamond hexagonal cross-section with a high inclination angle, where the slopes of the sides vary significantly and the cross-sectional asymmetry is prominent. This makes the steel shell positioning process susceptible to problems such as uneven distribution of support forces at multiple points, local buckling, and position coupling offsets.

[0003] Traditional construction methods rely on empirical trial adjustments and static support systems, making it difficult to dynamically adapt to changes in the steel shell's posture and load distribution. This can easily lead to defects such as cumulative deviations in the steel shell installation and stress concentration in the joint. Furthermore, existing processes lack a coordinated optimization mechanism for controlling the linear shape of complex sections and balancing the distribution of supporting forces. This makes it difficult to achieve an effective balance between micron-level posture correction and structural safety, severely restricting the construction quality and engineering reliability of these special-section steel-concrete joints.

[0004] Therefore, in order to meet the construction needs of large-angle diamond hexagonal cross-section steel-concrete joint sections, a systematic solution that integrates high-precision posture monitoring, dynamic support force optimization and closed-loop feedback control is urgently needed. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a construction method for the steel-concrete joint section of a hexagonal cross-section large-cutting-angle and inclination-angle combined tower, which solves the problem that traditional construction methods are difficult to simultaneously achieve high-precision positioning of the steel shell and dynamic balanced distribution of supporting force in the construction of complex-section steel-concrete joint sections, resulting in accumulated posture deviations and local stress concentration.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut angle and inclination combined tower comprises the following steps:

[0007] (a) A strain sensor group and laser calibration targets are deployed on the steel shell of the steel-concrete joint section, and a mapping relationship between the steel shell's position and posture is established based on the real-time collected strain data and laser calibration data;

[0008] (b) generating adjustment parameters of the three-way jack by a multi-objective optimization algorithm according to the deviation between the posture of the steel shell and the design posture, and driving the three-way jack to correct the posture of the steel shell;

[0009] (c) Based on the corrected steel shell posture and strain data, the target force value of each reinforcement support is calculated through a graph theory model, and the length of the retractable support rod is dynamically adjusted to match the target value;

[0010] (d) Based on the actual force data of the adjusted support rods, the weight parameters of the multi-objective optimization algorithm are fed back and corrected, and steps (b) to (c) are iteratively executed until the steel shell posture and the support rod forces meet the preset convergence conditions.

[0011] Preferably, step (a) comprises:

[0012] Strain sensor groups are arranged on the outer wall of the steel shell and at key stress points inside. The sensors are arranged in a hexagonal symmetrical distribution, with the spacing not exceeding one-quarter of the side length of the steel shell section;

[0013] Laser calibration targets are arranged symmetrically on the outer surface of the steel shell according to the vertices of the hexagonal section, and the distance between the target points does not exceed half of the side length of the steel shell section;

[0014] Real-time collection of strain data and laser calibration data to establish the mapping relationship between the steel shell posture parameters;

[0015] The mapping relationship of the steel shell posture is established through a joint calibration model, specifically including:

[0016] The strain sensor data matrix and the laser calibration target coordinate matrix are input into the nonlinear function to fit the real-time posture parameters of the steel shell.

[0017] Preferably, in step (b), the objective function of the multi-objective optimization algorithm is:

[0018] F(ΔX)=w1·||PP design / / 2+w2·Var(S)

[0019] Wherein, ΔX represents the adjustment amount of the three-way jack; P is the real-time posture parameter of the steel shell obtained through the mapping relationship in step (a); P design is the designed pose parameter; Var(S) represents the variance of the strain sensor group, and w1 and w2 are weight coefficients.

[0020] Preferably, the multi-objective optimization algorithm is solved by a non-dominated sorting genetic algorithm, and the constraints include the stroke limit of the three-way jack and the local deformation threshold of the steel shell.

[0021] Preferably, in step (c):

[0022] The graph theory model is a support force distribution model based on the maximum flow-minimum cut algorithm. Its input is the load matrix L generated by the steel shell strain data, and its output is the target force value F of each support rod. target ={f1,f2,…,fk};

[0023] The load matrix L=[l ij ] k×k The calculation method is:

[0024]

[0025] Among them, s i ,s j For adjacent support points v i and v j Corresponding strain sensor measurement value; A is the cross-sectional area of ​​the steel shell; k is the total number of reinforcement support points.

[0026] Preferably, the dynamically adjusting the length of the telescopic support rod comprises the following steps:

[0027] Coarse adjustment stage: All support rods are lifted synchronously at a constant speed until the contact pressure F contact ≥F threshold , where F threshold is a preset pressure threshold;

[0028] Fine-tuning stage: according to the target force value F target ={f1,f2,…,f k}, independently adjust the extension and contraction of each support rod ΔL={Δl1,Δl2,…,Δl k},satisfy:

[0029]

[0030] Among them, f′ i is the actual force on the i-th support rod; □ is the preset allowable ratio of force deviation; Δl i is the expansion and contraction of the i-th support rod.

[0031] Preferably, the retractable support rod is driven by a screw, and its retractable amount Δl i The rotation angle is controlled by the servo motor.

[0032] Preferably, in step (d), the feedback correction includes:

[0033] According to the actual force variance Var(F actual ) and target force variance Var(F target ) ratio, dynamically adjust the weight coefficient w2 of the strain variance term in the multi-objective optimization algorithm, and its correction formula is:

[0034]

[0035] Among them, α is the preset correction rate coefficient, w ′2 is the modified weight coefficient.

[0036] Preferably, the termination condition of the iterative execution is:

[0037] The steel shell posture deviation is less than the preset accuracy threshold;

[0038] The force variance of the support rod is less than the preset equilibrium threshold.

[0039] Preferably, the method further comprises the following steps:

[0040] Transport the steel shell of the steel-concrete composite section to the construction site and hoist it to the construction platform using a floating crane and a powered flatbed transport vehicle;

[0041] Pre-install the internal steel bars of the steel-concrete joint section on the construction platform;

[0042] Before pouring concrete in the front section of the steel-concrete joint section, install the steel shell fine-adjustment anchor piers and retractable support rods, and reserve a concrete section for pouring simultaneously with the steel shell of the steel-concrete joint section;

[0043] After the front section is poured, place the three-way jack on the steel shell fine-tuning anchor pier to fix it;

[0044] Hoist the steel shell of the steel-concrete combined section onto the three-way jack and use the three-way jack to fine-tune the position;

[0045] Use the retractable support rod to replace the support, remove the exposed part of the anchor pier, the three-way jack and the servo motor drive device of the retractable support rod;

[0046] Install the steel shell of the steel-concrete composite section and the internal steel bars of the reserved concrete section;

[0047] Install the reserved concrete section formwork. The upper edge of the formwork is fixed to the outer wall of the steel shell of the steel-concrete joint section through the tie rod lock nut, and the lower edge is pressed to the concrete surface of the previous section through the L-shaped pressing member;

[0048] A locking device is installed at the top of the steel shell in the steel-concrete joint section, and concrete is poured and maintained.

[0049] The invention provides a construction method for a steel-concrete combined section of a hexagonal cross-section large-cutting-angle and inclination-angle combined tower.

[0050] It has the following beneficial effects:

[0051] 1. The present invention constructs a posture parameter mapping model by fusing multi-source data from a strain sensor group and a laser calibration target, and combines it with a multi-objective optimization algorithm to correct the posture of the steel shell in real time, effectively solving the problem of balancing positioning accuracy and structural force in traditional methods. It ensures high-precision positioning of the steel shell in a complex construction environment while avoiding structural damage caused by local overload.

[0052] 2. This invention dynamically distributes support rod forces based on a maximum flow-minimum cut algorithm, transforming the strain distribution of the steel shell into the coordinated load-bearing requirements of the support system, significantly improving the rationality of the load transfer path. Through a phased control strategy of coarse and fine adjustments, stress concentration caused by asynchronous adjustment is avoided while achieving micron-level force balance, enhancing the stability and reliability of the temporary support system.

[0053] 3. This invention incorporates a feedback correction mechanism to dynamically adjust the optimization algorithm's weight coefficients based on the actual force variance of the support rods, enabling the system to adapt to changes in the steel shell's material properties and environmental interference. Combined with automatic fallback and reset logic in abnormal conditions, this creates a cross-level closed-loop fault-tolerant control system, effectively addressing uncertainties such as sensor noise and mechanical hysteresis.

[0054] 4. The present invention adopts an integrated design of a detachable servo motor drive and a permanently retained support rod body. After completing high-precision adjustment, the temporary drive module is quickly removed, and the retained support rod and steel shell reinforcement structure work together, which not only reduces the interference of temporary components on subsequent construction, but also improves the overall load-bearing performance of the permanent structure.

[0055] 5. This invention utilizes a modular quick-release interface and adaptive optimization algorithm to reduce manual intervention and shorten the iteration cycle for posture correction and support adjustment. Furthermore, the reusable design of the drive mechanism and the permanent retention of support rods reduce material loss, providing a cost-effective solution for large-scale steel structure projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Schematic diagram of the method flow of the present invention;

[0057] Figure 2 This is a schematic diagram of the steel shell transportation of the steel-concrete combined section in step S1 of the present invention;

[0058] Figure 3 This is a schematic diagram of the arrangement of the floating crane, powered flatbed transport vehicle, and construction platform of the present invention;

[0059] Figure 4 This is a schematic diagram of the arrangement before pouring concrete in the front section of the steel-concrete combined section in step S3 of the present invention;

[0060] Figure 5 This is a schematic diagram of the arrangement of the fulcrum and the template tie rod locking nut in step S4 of the present invention;

[0061] Figure 6 This is a schematic diagram of the process of hoisting the steel shell of the steel-concrete combined section onto a three-way jack in step S5 of the present invention;

[0062] Figure 7 This is a schematic diagram of the steel shell of the steel-concrete combined section in step S5 of the present invention being on a three-way jack;

[0063] Figure 8 This is a schematic diagram of step S6 of the present invention after the support is replaced using the telescopic support rod;

[0064] Figure 9 A schematic diagram after the concrete segment formwork is installed is reserved for step S8 of the present invention.

[0065] Among them, 1. Transport ship; 2. Steel shell of steel-concrete combined section; 3. Floating crane; 4. Hoist; 5. Powered flatbed transport vehicle; 6. Tower crane; 7. Construction platform; 8. Steel shell fine-tuning anchor pier; 9. Retractable support rod; 10. Three-way jack; 11. Reserved concrete section; 12. Local reinforcement of fulcrum; 13. Tie rod locking nut; 14. L-shaped die-cast component; 15. Locking device. DETAILED DESCRIPTION

[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0067] Please see the attached Figure 1 -Attached Figure 9 The present invention provides a construction method for the steel-concrete joint section of a hexagonal cross-section large-cutting-angle and inclination-angle combined tower, which realizes high-precision positioning of the steel shell and structural stability control based on the closed-loop control logic of the fusion of strain and laser calibration data.

[0068] like Figure 1 As shown, the construction method of the steel-concrete combined section of the hexagonal cross-section large-cut angle and inclination combined tower may include the following steps:

[0069] S1. Transport the steel shell 2 of the steel-concrete composite section to the construction site and hoist it to the construction platform 7 using a floating crane 3 and a powered flatbed transport vehicle 5.

[0070] S2, pre-installing the internal steel bars of the steel-concrete joint section steel shell 2 on the construction platform 7;

[0071] S3. Before pouring concrete in the front section of the steel-concrete composite section, install the steel shell fine-tuning anchor pier 8 and the retractable support rod 9, and reserve a concrete section for pouring the steel shell 2 of the steel-concrete composite section simultaneously. After pouring the front section, place the three-way jack 10 on the steel shell fine-tuning anchor pier 8 to fix it;

[0072] S4. When processing the steel shell 2 of the steel-concrete combined section, set the support point for local reinforcement 12 at the anchor pier support, and set the template tie rod locking nut 13 on the bottom outer wall plate;

[0073] S5. Hoist the steel shell 2 of the steel-concrete joint section onto the three-way jack 10 and perform fine adjustment of the position by the three-way jack 10;

[0074] S6. Use the telescopic support rod 9 to replace the support. After the replacement, remove the exposed part of the anchor pier, the three-way jack 10 and the servo motor drive device of the telescopic support rod 9;

[0075] S7, installing the steel shell 2 of the steel-concrete combined section and the internal steel bars of the reserved concrete section 11;

[0076] S8. Install the template for the reserved concrete section 11. The upper edge of the template is fixed to the outer wall of the steel shell 2 of the steel-concrete combined section by the tie rod locking nut 13, and the lower edge is pressed to the concrete surface of the previous section by the L-shaped pressing member 14;

[0077] S9. Install a locking device 15 at the top of the steel shell 2 in the steel-concrete joint section, pour concrete and maintain.

[0078] The following is a detailed description of each step in the method of the present invention, which comprehensively explains the specific implementation principles, technical details and processes of each step.

[0079] For step S1, Figure 2 and Figure 3 As shown, in this embodiment, the steel shell 2 of the steel-concrete composite section is transported to the construction site via a transport vessel 1. A floating crane 3 in conjunction with a dedicated lifting device 4 is used to lower it onto a powered flatbed transporter 5. The powered flatbed transporter 5 then transports the steel shell to the operating area of ​​a tower crane 6. The tower crane 6, in conjunction with the lifting device 4, hoists the steel shell onto a construction platform 7 for temporary storage. Anti-deformation brackets are pre-installed on the construction platform 7 to ensure cross-sectional geometric accuracy during the hoisting process.

[0080] In step S2, in this embodiment, the pre-binding of the internal steel shell 2 of the steel-concrete composite section is completed on the construction platform 7. The steel skeleton utilizes a modular, segmented design, with positioning fixtures controlling the spacing between the main bars and the thickness of the protective layer. The pre-embedded sleeves are precisely aligned with the welded anchor points on the inner wall of the steel shell. This step shifts the traditional high-altitude steel installation work on a tower to a ground-based platform, significantly reducing the risks of working at height.

[0081] For step S3, Figure 4 As shown, in this embodiment, before pouring concrete in the front section of the steel-concrete combined section, the steel shell fine-tuning anchor pier 8 and the telescopic support rod 9 system are installed:

[0082] Anchor piers: They are arranged symmetrically along the six vertices of the hexagonal cross-section of the bottom surface of the steel shell, and the pier foundation is embedded in the concrete of the previous section;

[0083] Telescopic support rod 9: driven by a planetary roller screw, with both ends of the support rod connected to the steel shell outer wall reinforcement node and the platform base embedded in the front section concrete through quick-release flanges;

[0084] Reserved concrete section 11: When pouring the previous section of concrete, reserve a 1.5m high concrete section. Its formwork is connected to the bottom of the steel shell through tongue and groove to ensure the density of the joint when it is poured synchronously with the steel shell later.

[0085] For step S4, Figure 5 As shown, in this embodiment, structural reinforcement is performed at key locations during the steel shell processing stage:

[0086] Local reinforcement of the support point 12: Weld stiffening ribs at the inner wall of the steel shell corresponding to the anchor pier. The thickness of the ribs is 1.2 times the thickness of the steel shell to form a local rigid support area.

[0087] Formwork tie rod locking nuts 13: High-strength alloy nuts are embedded at equal intervals in the outer wall of the steel shell bottom opening to secure the formwork tie rods. Rubber seals are embedded in the nuts to prevent concrete slurry from seeping into the threads.

[0088] For step S5, Figure 6 and 7 As shown, in this embodiment, the steel shell hoisting and posture fine adjustment are achieved through a closed-loop control process of real-time monitoring-optimization feedback-execution adjustment, which specifically includes:

[0089] In this embodiment, the deployment and data collection of the steel shell posture monitoring system are carried out in stages before the steel shell is hoisted onto the three-way jack 10. The specific implementation process is as follows:

[0090] After the steel shell is transported to the construction platform 7 and the internal steel bars are pre-installed, the strain sensor group is first deployed. The strain sensors are evenly spaced along each side of the hexagonal cross-section of the outer wall of the steel shell, with a spacing of one-quarter of the cross-section length (in the preferred embodiment, when the side length is 2m, the spacing is 0.5m). The installation position is preferably selected at the midpoint of the edge line and one-quarter of the points on both sides to cover the bending and shear strain gradient areas; at the same time, the internal sensors are arranged at the intersection of the internal reinforcement ribs and the annular stiffening plate of the steel shell, and are pre-installed using a welded fixed bracket. An insulating layer is set between the bracket and the steel shell base to isolate electrical noise interference. The outer wall sensor is directly attached with epoxy resin adhesive, and the surface is covered with waterproof sealant to improve durability.

[0091] On this basis, the laser calibration targets were installed and calibrated. The targets were symmetrically arranged at the six vertices of the hexagonal cross-section on the outer surface of the steel shell. The target body was a spherical marker made of high-reflectivity ceramic material with a diameter of 50 mm, fixed by adsorption on a magnetic base. Before installation, a total station was used to accurately lay out the target positions to ensure that the line connecting their centers and the geometric center of the cross-section was evenly distributed at 60°, and a three-dimensional spatial calibration network was formed at intervals of 1 m along the height direction of the steel shell. The initial coordinates of the targets were measured by the total station and stored in the reference database. The measurement error was controlled within ±0.1 mm, providing a spatial reference for subsequent position and posture calculations.

[0092] The data acquisition system was deployed simultaneously with sensor installation. The strain sensors were connected to the data acquisition system via shielded cables. The system, which integrates signal amplification and low-pass filtering, outputs strain data at a sampling rate of 100 Hz. The laser tracking system updates target coordinates at a 10 Hz frequency, synchronized with the strain data via timestamps. During acquisition, the strain data is smoothed using a Kalman filter to suppress ambient vibration noise, while the laser coordinate data is spatially interpolated to compensate for brief data gaps caused by occlusion.

[0093] It should be noted that the installation of sensors and targets is performed after the construction of the local reinforcement structure of the steel shell (such as the stiffener and the template locking nuts) is completed to avoid damage to the monitoring equipment from high welding temperatures or mechanical shock. After installation, the system is debugged and initially calibrated: multiple sets of strain-target data are collected under known pose disturbances. A nonlinear mapping function is fitted using a machine learning algorithm to establish a correlation model between the steel shell strain distribution, target displacement, and six-degree-of-freedom pose parameters, providing a data fusion foundation for subsequent closed-loop control.

[0094] In this embodiment, the mapping relationship modeling of the steel shell posture parameters is based on the fusion of multi-source sensor data and machine learning algorithm. The specific process is as follows:

[0095] First, the input data of the pose solution model includes the measurement value matrix S = [s1, s2, ..., s n ] T And the coordinate matrix of the laser calibration target D=[d1,d2,…,d m ](where d j =[x j ,y j ,z j ] T is the real-time coordinate of the j-th target).

[0096] To establish the strain, displacement and six-degree-of-freedom posture parameters P = [x, y, z, θ x ,θ y ,θ z ] TIn order to avoid the nonlinear correlation, a mapping function P = f(S, D) needs to be constructed through pre-calibration experiments.

[0097] Furthermore, a neural network algorithm is used to train the mapping function f. The network input layer is the joint feature vector of strain data and target coordinates, and the specific construction method is:

[0098] Normalize S and D and concatenate them into in is the normalized result after subtracting the baseline value from the strain data. is the offset of the target coordinate relative to the initial position.

[0099] The hidden layer of the network contains a 3-layer fully connected structure, the activation function uses ReLU, and the output layer directly regresses the six-degree-of-freedom posture parameter P.

[0100] During training, the mean square error (MSE) is used as the loss function, and the network weights are optimized through back propagation, ultimately obtaining a pose solution model that can decouple rigid body motion and elastic deformation.

[0101] It should be noted that the generalization ability of the model is ensured through cross-validation: the dataset is divided into training set and test set in a ratio of 7:3, and the error of the test set is controlled within the position translation component ≤ 0.5 mm and the rotation component ≤ 0.01°.

[0102] Due to the symmetry characteristics of the hexagonal section, the training data needs to cover all possible symmetry transformation combinations (such as rotating 60°, 120° around the center axis of the section, etc.) to improve the model's adaptability to the complex postures of the steel shell.

[0103] In the real-time pose solution stage, the collected strain data S and target coordinates D are pre-processed and input into the trained neural network model to directly output the pose parameter estimation value P.

[0104] To cope with time-varying interference in the construction environment (such as temperature drift and mechanical vibration), the model adopts a sliding window incremental update strategy: every 10 minutes, the latest data is added to the training set and the network weights are fine-tuned to enable the model to dynamically adapt to the slow changes in the steel shell material properties and boundary conditions.

[0105] In addition, the model output results must pass physical rationality verification: if the predicted value of the posture parameter P causes the strain distribution S to exceed the elastic limit of the material (calculated according to Hooke's law), or the target displacement D conflicts with the spatial kinematic constraints of the posture parameter, it is judged as an abnormal output, triggering a system alarm and suspending the adjustment operation.

[0106] In summary, through pre-calibrated data-driven modeling, dynamic update and anomaly verification mechanism, the pose mapping relationship can accurately characterize the spatial motion characteristics of the steel shell under large cutting angle and tilt angle combinations.

[0107] In this embodiment, the posture correction decision is made by balancing the geometric accuracy of the steel shell and the structural safety requirements through a multi-objective optimization algorithm. The specific implementation process is as follows:

[0108] First, a multi-objective optimization function including posture deviation and strain balance is constructed. The objective function is defined as:

[0109] F(ΔX)=w1·||PP design / / 2+w2·Var(S)

[0110] Where ΔX = [Δx, Δy, Δz] T is the adjustment amount of the three-way jack, P is the steel shell posture parameter calculated in real time through sensor data, and P design To design the posture, represents the variance of the strain sensor group, w1 and w2 are dynamic weight coefficients.

[0111] On this basis, the non-dominated sorting genetic algorithm (NSGA-II) is used to solve the optimization problem.

[0112] When the algorithm is initialized, a candidate solution set containing 100 individuals is randomly generated, and each individual corresponds to a set of jack adjustment values ​​ΔX. The individual fitness is evaluated by the objective function value F(ΔX) and the degree of constraint violation. The constraints include:

[0113] Jack travel limit: single-axis adjustment Δx∈[-10mm,10mm] to avoid mechanical overtravel;

[0114] Local deformation threshold of steel shell: local bending angle |θ x |≤0.01rad, to prevent plastic deformation;

[0115] Pose convergence boundary: the final pose deviation must satisfy ||PP design / / ≤2mm.

[0116] During the iterative optimization process, the offspring population is generated by simulating binary crossover and polynomial mutation operations, with the crossover probability set to 0.9 and the mutation probability set to 0.1. Each generation of the population is divided into levels by fast non-dominated sorting, and the crowding degree is calculated to maintain the diversity of solutions. After 50 generations of iteration, the optimal solution ΔX that takes into account both posture accuracy (dominated by w1) and strain balance (dominated by w2) is selected from the Pareto front. * .

[0117] Furthermore, a penalty function approach is used to address constraints: for solutions that exceed travel or deformation thresholds, a penalty term λ·violation is added to their fitness, where λ is a penalty factor (set to 1e5), forcing the algorithm to prioritize satisfying hard constraints. Furthermore, the algorithm monitors steel shell strain data in real time during runtime. If an adjustment causes Var(S) to increase beyond a safety threshold (e.g., 200% of the baseline value), the optimization is immediately terminated and manual intervention is triggered.

[0118] After the decision is made, the three-way jack 10 is adjusted according to the optimal adjustment value ΔX * Perform posture correction. During the execution process, the laser tracker provides real-time feedback on the actual displacement of the steel shell and compares it with the expected adjustment. If the deviation exceeds 5%, the optimization calculation is re-triggered to form a closed-loop control loop. Finally, the termination condition of posture correction is that ||PP design / / The change is less than 0.1mm to ensure convergence stability.

[0119] In summary, through the multi-objective optimization algorithm and dynamic weight mechanism, the posture correction decision can achieve the coordinated optimization of steel shell positioning accuracy and structural safety under complex constraints.

[0120] For step S6, Figure 8 As shown, in this embodiment, the retractable support rod 9 is replaced by a mechanical distribution model and closed-loop feedback control. The specific implementation process is as follows:

[0121] In this embodiment, the construction of the support force distribution model and the load matrix calculation are achieved by integrating the steel shell strain data with the graph theory optimization algorithm. The specific implementation process is as follows:

[0122] First, the load matrix L=[l ij ] k×k The construction of is based on the real-time data of the strain sensor group deployed in step S5. ij Characterize adjacent support points v i and v j The load transfer relationship between them is calculated as follows:

[0123]

[0124] Among them, s i 、s j Arranged at the support point v i 、v jThe strain sensor measurements near the support reflect the local load distribution of the steel shell under the combination of large cut angles and inclinations of the hexagonal cross section. A is the cross-sectional area of ​​the steel shell and is used to convert the strain value into an equivalent axial force. This calculation method eliminates single-point measurement noise by averaging the strains of adjacent support points and, combined with cross-sectional geometric properties, links microscopic strain to macroscopic mechanical response, providing a quantitative basis for support force distribution.

[0125] On this basis, the load matrix is ​​input into the graph theory model based on the maximum flow-minimum cut algorithm. The model is based on the support points v1, v2, ..., v k As a node, with l ij A directed graph is constructed for the edge weights, where the direction of the edge weights is determined by the load transfer path (e.g., from high strain areas to low strain areas).

[0126] By solving the maximum flow-minimum cut problem, the target force value F of each support rod is determined. target ={f1,f2,…,f k}, maximizing the force transmission efficiency of the overall support system while satisfying the following constraints:

[0127] 1. Load balance: The force distribution of the support rods must match the local strain distribution of the steel shell to avoid overloading the support rods in low-load areas;

[0128] 2. Bearing capacity limitation: The load on a single support rod must not exceed its design threshold to prevent structural failure;

[0129] 3. Path Optimization: Prioritize the distribution of loads to adjacent support rods to reduce bending losses in the force transmission path.

[0130] It should be noted that this model is closely related to the posture closed-loop control logic of step S5:

[0131] The deformation and displacement of the steel shell during the posture correction process will change the strain distribution, so the load matrix needs to be dynamically updated.

[0132] After each posture adjustment is completed, the strain data is recollected and L is updated to ensure that the support force distribution model responds to the changes in the steel shell state in real time.

[0133] In addition, the target force value F output by the model target As input to the support rod fine tuning phase in step S6, the independent adjustment of the telescopic support rods is guided.

[0134] Furthermore, the construction of the load matrix depends on the preset local reinforcement structure of the steel shell in step S4: the support point v iThe position of the strain sensors is aligned with the weld area of ​​the steel shell stiffeners, ensuring that the supporting force is effectively transmitted to the main steel shell through the reinforced structure, avoiding local buckling. Furthermore, the spacing of the strain sensors (one-quarter of the cross-section length) matches the spacing of the support points. This allows the calculation of the mean strain between adjacent support points to cover the bending and shear deformation gradients of the steel shell, improving the physical rationality of the load matrix.

[0135] In summary, through the synergistic effect of the dynamic construction of the load matrix and the graph theory optimization model, the support force distribution process can accurately match the mechanical requirements of the steel shell under complex deformation states.

[0136] In this embodiment, the dynamic adjustment process of the telescopic support rod is precisely matched with the load distribution of the steel shell through a staged control strategy. The specific implementation process is as follows:

[0137] First, during the rough adjustment phase, all support rods are synchronously raised at a constant speed. The goal of this phase is to establish initial contact pressure between the support rods and the steel shell to avoid local stress concentration caused by asynchronous lifting.

[0138] The lifting speed is set according to the mechanical response characteristics of the support system. Preferably, the speed value should ensure that the impact force of the support rod at the moment of contact is lower than the elastic limit of the steel shell material.

[0139] During the synchronous lifting process, the contact pressure F of each support rod is monitored in real time through the built-in pressure sensor contact , when the F of all support rods contact ≥F threshold When , it is determined that the coarse adjustment is completed.

[0140] Threshold F threshold The contact pressure is determined based on the steel shell's deadweight distribution and construction load calculation to ensure that it is sufficient to maintain the temporary stability of the steel shell while avoiding plastic deformation caused by excessive pre-tightening.

[0141] On this basis, the fine-tuning phase is entered. According to the target force value F output by the maximum flow-minimum cut algorithm in step S6 target ={f1,f2,…,f k}, independently adjust the extension and contraction amount Δl of each support rod i .

[0142] During adjustment, a servo motor drives a planetary roller screw mechanism, converting the motor's rotation angle into linear displacement. Optimally, the screw pitch and reduction ratio are designed to achieve micron-level resolution for each rotation angle adjustment, meeting high-precision adjustment requirements.

[0143] Adjustment amount of each support rod Δl i Dynamic calculation based on actual force deviation:

[0144]

[0145] Among them, K p With K i is the proportional-integral control parameter, calibrated according to the stiffness characteristics of the support rod; f′ i The actual force value fed back by the pressure sensor. The adjustment termination condition is that the force deviation of all support rods meets Ensure that the overall force of the support system is balanced.

[0146] It should be noted that the independence adjustment during the fine-tuning phase relies on the closed-loop posture control architecture established in step S5: real-time monitoring data of the steel shell posture parameter P is used to correct the support rod adjustment amount to prevent secondary displacement of the steel shell due to changes in local support force. For example, if the translation component of P exceeds the limit due to adjustment of a support rod, the rod adjustment is immediately suspended and the posture correction iteration is triggered, forming a cross-validation mechanism between support adjustment and posture control.

[0147] For abnormal working conditions, the force mutation and position deviation of the support rod are monitored in real time during the adjustment process. If the actual force f′ of a support rod is i After adjustment, the deviation from the target value exceeds the allowable ratio □, or the steel shell posture deviation || PP design / / If the convergence threshold is exceeded, it is considered an abnormal state. At this time, the system automatically rolls back to the last stable configuration and re-executes the posture optimization in step S5 and the support force distribution calculation in step S6 until the abnormality is eliminated.

[0148] In summary, through the staged control of coarse adjustment and fine adjustment, the high-precision execution of servo drive, and the coordinated verification with the above steps, the dynamic adjustment process of the support rod can achieve the dual goals of force balance and posture stability of the steel shell in a complex construction environment.

[0149] In this embodiment, the dynamic adjustment of the weight coefficient is achieved through a feedback closed loop of the support rod force data. The specific implementation process is as follows:

[0150] First, after each support rod fine-tuning is completed, the actual force variance is calculated in real time Force variance with target in and are the mean of actual and target forces respectively. Variance ratio It reflects the degree of deviation from the balance of force distribution between support rods. When R>1, it indicates that the actual force fluctuation is greater than expected, and the weight of the strain variance term in the multi-objective optimization algorithm needs to be increased to improve the balance.

[0151] On this basis, the weight coefficient w2 is dynamically modified according to the variance ratio:

[0152] w′2=w2·(1+α·R)

[0153] Here, α is the preset correction rate coefficient, which controls the sensitivity of the weight adjustment. The corrected weight coefficient w′2 is normalized (w′1 = 1 - w′2) and used in the next round of multi-objective optimization calculations in step S5. For example, when the actual force variance is significantly higher than the target value, w2 is increased to strengthen the weight of strain equilibrium in the optimization objective, prioritizing the reduction of stress nonuniformity on the steel shell during subsequent posture correction.

[0154] It should be noted that the weight correction mechanism is deeply coupled with the pose optimization algorithm in step S5: the corrected w′1 and w′2 directly act on the objective function F(ΔX)=w′1·||PP design / / 2 + w′²·Var(S), enabling the optimization algorithm to adaptively balance pose accuracy and force balance based on the real-time state of the support system. This logic forms a closed-loop control chain of "support adjustment → feedback correction → optimization weight update → pose recalibration," ensuring continuous optimization of system performance over multiple iterations.

[0155] Furthermore, the value of the correction rate coefficient α matches the convergence rate of the support rod adjustment in step S6. Preferably, α is dynamically adjusted based on historical iteration data: if the rate of change of R in two consecutive iterations is below a threshold, α is increased to accelerate weight adjustment; conversely, if R fluctuates dramatically, α is decreased to avoid overshoot. This adaptive strategy avoids manual intervention and improves system stability under complex operating conditions.

[0156] For abnormal working conditions, when R continuously exceeds the safety threshold (e.g., R ≥ 3), the support system is judged to be unstable and the abnormal handling process is automatically triggered:

[0157] The current iteration is paused, the weight coefficient is returned to the previous stable state, and the support force distribution and adjustment operation of step S6 is re-executed.

[0158] If convergence is still not possible after backing off, the posture correction algorithm of step S5 is started to reset the global posture of the steel shell, thereby forming a cross-level fault tolerance mechanism with the above steps.

[0159] In summary, through the real-time feedback of the variance ratio and the dynamic correction of the weight coefficient, the multi-objective optimization algorithm can adapt to the changes in the stress state of the steel shell during the iteration process, ensuring the continuous optimization of the correction accuracy and structural safety.

[0160] In this embodiment, the determination of the completion of the support replacement and the removal of the drive device are achieved through multi-index verification and modular quick-disassembly design. The specific implementation process is as follows:

[0161] First, the following conditions must be met simultaneously to determine whether the brace replacement operation is complete:

[0162] 1. Posture convergence condition: The steel shell posture parameter P calculated in real time by the posture monitoring system in step S5 satisfies ‖PP design ‖≤δ p , where δ p is the preset pose accuracy threshold;

[0163] 2. Force balance condition: actual force variance Var(F actual )≤δ f , δ f The equilibrium threshold is set according to the material properties of the steel shell and the design specifications;

[0164] 3. Stability condition: The fluctuation amplitude of posture deviation and force variance in three consecutive iterations is less than 5%, ensuring that the system enters a steady state.

[0165] Once the above conditions are met, the drive unit is dismantled. This process is divided into two stages: first, disconnecting the servo motor from the lead screw drive mechanism, followed by removing the exposed portion of the temporary anchoring pier. The servo motor is coupled to the lead screw end via a quick-release flange, which is secured with circumferentially distributed anti-loosening bolts. During dismantling, the bolts are loosened in a symmetrical sequence to prevent the lead screw from shifting due to sudden stress changes. A laser tracker monitors the displacement of the support rod end to ensure the stability of the steel shell during dismantling.

[0166] Next, the three-way jack 10 and the temporary anchor pier are removed. After the three-way jack 10 is removed, the temporary anchor pier is removed using a hydraulic shear to cut the exposed rebar, leaving the anchor section embedded in the concrete. The cutting position is determined based on the embedded mark in step S3, ensuring that the remaining anchor length meets the pullout resistance requirements of the permanent structure.

[0167] It should be noted that after the drive unit is removed, the screw and sleeve of the telescopic support rod 9 remain as a permanent reinforcement structure. The screw position is fixed by a mechanical locking device. The locking force is set according to the final force value in the fine-tuning stage of step S6 to ensure that the support rod maintains the preset prestress when not driven.

[0168] In summary, through multi-index criteria, modular quick-disassembly design and locking retention mechanism, the support replacement operation can be completed efficiently while ensuring structural safety.

[0169] Regarding step S7, in this embodiment, the steel shell 2 of the steel-concrete combined section and the internal steel bars of the reserved concrete section 11 are connected by a mechanical sleeve, the axis of the sleeve is aligned with the pre-installed steel bars and then tightened, and at the same time, an anti-cracking steel mesh is laid in the joint area.

[0170] For step S8, the upper edge of the template of the reserved concrete section 11 is fixed to the locking nut of the outer wall of the steel shell through a double-headed pull rod; the lower edge is pressed to the surface of the front section concrete through an L-shaped pressure mold component 14, and the contact surface between the pressure mold and the concrete is filled with elastic sealant to prevent leakage during pouring.

[0171] For step S9, Figure 9 As shown, the locking device 15 is an adjustable annular structure, and the gap between the inner diameter of the locking mouth and the outer wall of the top of the steel shell is adjusted by bolts; the concrete is poured in layers, and after being vibrated and compacted, it is covered with a double layer of geotextile for moisture retention and maintenance.

[0172] In summary, the present invention deploys strain sensor groups and laser calibration targets on the outer wall and at key points within the steel shell, establishing a posture parameter mapping model through real-time data fusion. A load matrix is ​​constructed based on the strain data, and the maximum flow-minimum cut algorithm is used to solve the target force value of the support rods, guiding the staged adjustment of the support rods: in the coarse adjustment stage, synchronous lifting is used to establish contact pressure, and in the fine adjustment stage, independent adjustment is made according to the target value and feedback correction is introduced to dynamically optimize the weight coefficient. Finally, after achieving the posture accuracy and force balance threshold, the drive device is removed, and the main support rod body is retained as a permanent reinforcement structure. This method achieves the unity of high-precision positioning of the steel shell and structural safety through the coordinated fusion of multi-source sensor data, graph optimization, and closed-loop control.

[0173] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut angle and inclination combined tower, characterized in that: The following steps are involved: (a) A strain sensor group and laser calibration targets are deployed on the steel shell of the steel-concrete joint section, and a mapping relationship between the steel shell's position and posture is established based on the real-time collected strain data and laser calibration data; (b) generating adjustment parameters of the three-way jack by a multi-objective optimization algorithm according to the deviation between the posture of the steel shell and the design posture, and driving the three-way jack to correct the posture of the steel shell; (c) Based on the corrected steel shell posture and strain data, the target force value of each reinforcement support is calculated through a graph theory model, and the length of the retractable support rod is dynamically adjusted to match the target value; (d) Based on the actual force data of the adjusted support rods, the weight parameters of the multi-objective optimization algorithm are fed back and corrected, and steps (b) to (c) are iteratively executed until the steel shell posture and the support rod forces meet the preset convergence conditions.

2. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut angle and inclination combined tower according to claim 1 is characterized in that: Step (a) comprises: Strain sensor groups are arranged on the outer wall of the steel shell and at key stress points inside. The sensors are arranged in a hexagonal symmetrical distribution, with the spacing not exceeding one-quarter of the side length of the steel shell section; Laser calibration targets are arranged symmetrically on the outer surface of the steel shell according to the vertices of the hexagonal section, and the distance between the target points does not exceed half of the side length of the steel shell section; Real-time collection of strain data and laser calibration data to establish the mapping relationship between the steel shell posture parameters; The mapping relationship of the steel shell posture is established through a joint calibration model, specifically including: The strain sensor data matrix and the laser calibration target coordinate matrix are input into the nonlinear function to fit the real-time posture parameters of the steel shell.

3. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut angle and inclination combined tower according to claim 1 is characterized in that: In step (b), the objective function of the multi-objective optimization algorithm is: F(ΔX)=w1·||PP design / / 2+w2·Var(S) Wherein, ΔX represents the adjustment amount of the three-way jack; P is the real-time posture parameter of the steel shell obtained through the mapping relationship in step (a); P design is the designed pose parameter; Var(S) represents the variance of the strain sensor group, and w1 and w2 are weight coefficients.

4. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut and inclination combined tower according to claim 3 is characterized in that: The multi-objective optimization algorithm is solved by a non-dominated sorting genetic algorithm, and the constraints include the travel limit of the three-way jack and the local deformation threshold of the steel shell.

5. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut angle and inclination combined tower according to claim 1 is characterized in that: In step (c): The graph theory model is a support force distribution model based on the maximum flow-minimum cut algorithm. Its input is the load matrix L generated by the steel shell strain data, and its output is the target force value F of each support rod. target ={f1,f2,…,f k }; The load matrix L=[l ij ] k×k The calculation method is: Among them, s i ,s j For adjacent support points v i and v j Corresponding strain sensor measurement value; A is the cross-sectional area of ​​the steel shell; k is the total number of reinforcement support points.

6. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut and inclined combined tower according to claim 5, characterized in that: The dynamic adjustment of the length of the telescopic support rod comprises the following steps: Coarse adjustment stage: All support rods are lifted synchronously at a constant speed until the contact pressure F contact ≥F threshold , where F threshold is a preset pressure threshold; Fine-tuning stage: according to the target force value F target ={f1,f2,…,f k }, independently adjust the extension and contraction of each support rod ΔL={Δl1,Δl2,…,Δl k },satisfy: Among them, f′ i is the actual force on the i-th support rod; □ is the preset allowable ratio of force deviation; Δl i is the expansion and contraction of the i-th support rod.

7. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut and inclined combined tower according to claim 6, characterized in that: The telescopic support rod is driven by a screw, and its telescopic amount Δl i The rotation angle is controlled by the servo motor.

8. The method for constructing a steel-concrete joint section of a hexagonal cross-section large chamfer and inclination combined tower according to claim 1, characterized in that: In step (d), the feedback correction includes: According to the actual force variance Var(F actual ) and target force variance Var(F target ) ratio, dynamically adjust the weight coefficient w2 of the strain variance term in the multi-objective optimization algorithm, and its correction formula is: Wherein, α is the preset correction rate coefficient, and w′2 is the corrected weight coefficient.

9. The method for constructing a steel-concrete combined section of a hexagonal cross-section large-cut and inclined combined tower according to claim 8, characterized in that: The termination condition of the iterative execution is: The steel shell posture deviation is less than the preset accuracy threshold; The force variance of the support rod is less than the preset equilibrium threshold.

10. The method for constructing a steel-concrete joint section of a hexagonal cross-section large chamfer and inclination combined tower according to any one of claims 1 to 9, characterized in that: The following steps are also included: Transport the steel shell of the steel-concrete composite section to the construction site and hoist it to the construction platform using a floating crane and a powered flatbed transport vehicle; Pre-install the internal steel bars of the steel-concrete joint section on the construction platform; Before pouring concrete in the front section of the steel-concrete joint section, install the steel shell fine-adjustment anchor piers and retractable support rods, and reserve a concrete section for pouring simultaneously with the steel shell of the steel-concrete joint section; After the front section is poured, place the three-way jack on the steel shell fine-tuning anchor pier to fix it; Hoist the steel shell of the steel-concrete combined section onto the three-way jack and use the three-way jack to fine-tune the position; Use the retractable support rod to replace the support, remove the exposed part of the anchor pier, the three-way jack and the servo motor drive device of the retractable support rod; Install the steel shell of the steel-concrete composite section and the internal steel bars of the reserved concrete section; Install the reserved concrete section formwork. The upper edge of the formwork is fixed to the outer wall of the steel shell of the steel-concrete joint section through the tie rod lock nut, and the lower edge is pressed to the concrete surface of the previous section through the L-shaped pressing member; A locking device is installed at the top of the steel shell in the steel-concrete joint section, and concrete is poured and maintained.

Images