Intelligent measurement method for prestressed sag

Abstract

This invention discloses an intelligent method for measuring prestressed sag height, comprising: constructing a calibrable control space model based on acquired standardized design parameter data, including preset coordinate system calibrable parameters and prestressed curve shape calibrable parameters; collecting multi-point observation data at predetermined locations on the construction site, including spatial coordinate information of each measuring point; combining the multi-point observation data with the calibrable control space model, and by minimizing the residual between the theoretical sag height and the measured sag height, inverting and solving for the coordinate system calibrable parameters and the prestressed curve shape calibrable parameters to generate an updated self-calibrating control space model; and calculating the self-calibrating sag height deviation data corresponding to the multi-point observation data based on the self-calibrating control space model. This invention, through a two-way driven closed loop of model and data, realizes the transformation from static control to adaptive control, effectively eliminating systematic deviation interference and improving the measurement accuracy and risk control capability of prestressed construction.

Description

Technical Field

[0001] This invention belongs to the field of engineering surveying, and in particular to a method for intelligent measurement of prestressed sag. Background Technology

[0002] In the construction of prestressed concrete structures, the alignment accuracy of the prestressed corrugated pipes directly determines the stress state and structural safety of the completed bridge. Therefore, the accurate measurement and control of the prestressing sag is of significant engineering research importance. Precise sag control not only ensures that the prestressing tendons generate the equivalent load that meets the design specifications, but also avoids the risk of insufficient concrete cover thickness or structural cracking due to pipe positioning deviations, making it a key link in ensuring bridge durability.

[0003] Currently, the measurement and control of prestressed sag mainly relies on static layout methods based on design drawings. Construction workers typically establish a fixed engineering coordinate system according to the design documents and use a total station or level to lay out and verify points one by one according to the C4 curve equation or discrete coordinate points given in the design. During the acceptance phase, the vertical distance between the duct and the bottom formwork is directly measured and compared with the theoretical design value to determine whether the construction quality is up to standard. This method assumes that the geometry of the beam on site is completely consistent with the design model and does not consider the structural deformation factors during construction.

[0004] The existing technologies face the following pressing technical challenges: The core issue lies in the mismatch between the rigid design model and the time-varying physical state, leading to a coupling of systematic structural deviations and localized construction errors, making it impossible to accurately evaluate construction quality. Specifically, during actual construction, the actual spatial orientation of the beam often deviates systematically from the design coordinate system due to factors such as support settlement, formwork installation errors, or overall beam rotation. Existing methods lack the ability to reverse-correct the benchmark model based on measured data, leading to the misjudgment of this systematic overall offset as a large-area positioning error of the corrugated pipe, causing unnecessary rework or adjustments. Furthermore, existing methods only focus on deterministic numerical deviations, ignoring the cumulative propagation effect of error sources such as slab thickness, beam height, and measurement noise. This results in insufficient observation density in error-sensitive high-risk areas, making it difficult to effectively identify substantial quality defects masked by uncertainty. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent method for measuring prestressed sag height, so as to solve the above-mentioned problems existing in the prior art.

[0006] Technical solution, intelligent measurement method for prestressed sag, including:

[0007] Based on the acquired standardized design parameter data, a calibrable control space model is constructed, which includes preset calibrable parameters of the coordinate system and calibrable parameters of the prestress curve shape.

[0008] Collect multi-point observation data at predetermined locations on the construction site, including the spatial coordinate information of each measuring point;

[0009] By combining multi-point observation data with a calibrable control space model, and minimizing the residual between theoretical and measured sag, the calibrable parameters of the coordinate system and the calibrable parameters of the prestress curve shape are inverted and solved to generate an updated self-calibrating control space model.

[0010] Based on the self-calibration control space model, the self-calibration elevation deviation data corresponding to multi-point observation data are calculated.

[0011] Beneficial effects: This invention achieves the transformation from static control to adaptive control through a two-way driven closed loop of model and data, effectively eliminating systematic deviation interference and improving the measurement accuracy and risk control capability of prestressed construction. Attached Figure Description

[0012] Figure 1 A flowchart illustrating the steps of an intelligent method for measuring prestressed sag height provided in this application embodiment.

[0013] Figure 2 A flowchart illustrating the steps for constructing a calibrable controllable space model, as provided in an embodiment of this application.

[0014] Figure 3 A flowchart illustrating the steps for collecting multi-point observation data as provided in this application embodiment.

[0015] Figure 4 A flowchart illustrating the steps for generating an updated self-calibrating control space model, as provided in this application embodiment. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] It should be noted that the terms include and have, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0018] like Figure 1As shown, a smart method for measuring prestressed sag includes the following steps:

[0019] Based on the acquired standardized design parameter data, a calibrable control space model is constructed. The calibrable control space model includes preset coordinate system calibrable parameters and prestress curve shape calibrable parameters.

[0020] In other words, obtain standardized design parameter data, and based on the standardized design parameter data, construct a calibrable control space model that includes preset calibrable parameters of the coordinate system and calibrable parameters of the prestress curve shape.

[0021] In this embodiment, standardized design parameter data refers to basic data extracted from design drawings, structural calculation sheets, or engineering specifications and processed using a standardized format. Specifically, this data may include geometric dimensions of beam spans such as span length, beam height, and slab thickness; prestressing curve type information such as parabolic parameters and circular curve radii; and prestressing design sag information. The calibrable controllable space model is a dynamic parametric model built upon these static design data. Unlike traditional fixed models, this model introduces specific variable parameters, enabling it to be adjusted based on actual measurement results.

[0022] Specifically, calibrable parameters for the coordinate system can include variables describing the rotation angle and translation offset of the local coordinate system relative to the global engineering coordinate system. Calibrable parameters for the prestressed curve shape can include deviation factors describing the actual shape of the prestressed tendon relative to the designed shape, such as the control point elevation correction and end slope correction within the curve segment. By presetting these parameters, the system can subsequently use numerical optimization algorithms to find the parameter combination that best matches the actual physical state, overcoming systematic deviations caused by construction errors and structural deformation.

[0023] Multi-point observation data is collected at predetermined locations on the construction site, and the multi-point observation data includes the spatial coordinate information of each measuring point.

[0024] In other words, multi-point observation data containing spatial coordinate information of each measuring point is collected at predetermined locations on the construction site.

[0025] In this embodiment, unlike traditional single-point measurements, multiple representative locations along the beam's length are selected for measurement at the construction site. The selection of these locations should adequately reflect structural deformation and curve characteristics, such as key sections like mid-span, quarter-span, beam ends, and inflection points of the prestressed curve. The data acquisition process can be performed using a high-precision total station, scanner, or photogrammetric equipment. For each measurement point, the equipment acquires its three-dimensional spatial coordinates (X, Y, Z coordinates) in a unified engineering coordinate system. To ensure data validity, auxiliary information such as instrument height and prism height should be recorded during the acquisition process for necessary geometric corrections. These discrete sets of spatial coordinate points form the observational basis for subsequent parameter inversion; their quantity and distribution quality directly affect the accuracy of model calibration.

[0026] By combining multi-point observation data with a calibrable control space model, and minimizing the residual between the theoretical and measured sags, the calibrable parameters of the coordinate system and the calibrable parameters of the prestress curve shape are obtained through inversion, thereby generating an updated self-calibrating control space model.

[0027] Specifically, the system calculates the theoretical sag corresponding to each measuring point location based on the current model parameters, and simultaneously calculates the measured sag based on the observation data. An optimization objective function is constructed, which uses the sum of the squares of the differences between the theoretical and measured sags at all measuring points as an index to quantify the degree of agreement between the model and reality. Optionally, by applying numerical optimization techniques such as nonlinear least squares, Gauss-Newton's method, or Levenberg-Marquardt algorithm, the optimal parameter values ​​are iteratively searched to minimize the objective function. This set of optimal parameters represents the most realistic state of the coordinate system and curve shape. Using this set of parameters to update the original calibrable control space model yields the self-calibrating control space model. This model no longer relies solely on theoretical design values ​​but integrates on-site measured information, enabling a more accurate description of the current construction status.

[0028] Based on the self-calibration control space model, the self-calibration sag deviation data corresponding to multi-point observation data are calculated as a basis for evaluating construction quality.

[0029] In this embodiment, the self-calibrated sag deviation data serves as the final decision-making basis for quality control. Specifically, the system utilizes the corrected coordinate system in the updated self-calibrated control space model to remap the coordinates of the original observation points to the local reference system, calculating the corrected measured sag. Simultaneously, the target design sag at the corresponding location is calculated using the corrected prestressing curve equation. The difference between the two is the self-calibrated sag deviation. This eliminates spurious deviations caused by coordinate system establishment errors or systematic structural deformations, accurately reflecting the local construction errors of the corrugated pipe or prestressing tendons relative to the ideal alignment. Construction personnel can use the deviation data to determine whether adjustments to the support height or corrugated pipe position are necessary, or to conduct project acceptance. Compared to directly comparing with the original design values, this effectively distinguishes between systematic errors and random construction errors, avoiding misjudgments and unnecessary rework, and improving the scientific nature and efficiency of construction control.

[0030] This embodiment constructs a parameterized control model and uses multi-point field observation data to reverse-correct the model parameters, thereby achieving closed-loop control from static design to dynamic adaptation.

[0031] like Figure 2 As shown, in one exemplary embodiment, constructing a calibrable control space model includes:

[0032] Based on standardized design parameter data, the geometric information of the beam span and the structural dimension information are analyzed to construct local plane coordinate system data and initial beam bottom reference plane data that describe the geometric relationship of the beam foundation.

[0033] In this embodiment, constructing local planar coordinate system data is the first step in achieving spatial positioning. Specifically, the system extracts key physical feature points from standardized design parameter data. For example, the corner point at the intersection of the floor slab and beam can be selected as the reference point for station construction, denoted as P. corner At the same time, identify the design starting point of the prestressing tendon, denoted as P. start To establish a coordinate system that conforms to engineering practices, the prestressing starting point P is... start Define the origin O of the local coordinate system. Extract the design position P of the prestressing endpoint. end A unit vector is constructed pointing from the starting point to the ending point, and this vector is defined as the first axis of the local planar coordinate system, i.e., the X-axis. According to the right-hand rule or engineering convention, the direction perpendicular to the first axis and parallel to the floor slab plane is defined as the second axis, i.e., the Y-axis. The planar positional relationships of the beams are then established. Based on this, the system, combined with the design beam height H, is then... beam and design plate thickness T slab Based on structural dimension information, determine the vertical position of the beam bottom reference plane, construct initial beam bottom reference plane data, and form a complete initial geometric framework.

[0034] Construct the plane rotation angle and origin translation offset of the local plane coordinate system data relative to the preset engineering measurement coordinate system, and mark the plane rotation angle and origin translation offset as coordinate system calibrable parameter data.

[0035] In this embodiment, specific geometric transformation parameters are introduced to capture the actual deviation of the coordinate system during on-site construction. Specifically, an angle variable, denoted as Δθ, is defined to describe the rotation of the local coordinate system relative to the global engineering coordinate system around the Z-axis, reflecting the actual deflection of the beam axis in the horizontal plane. Simultaneously, the translational offsets of the origin in the three coordinate axis directions are defined, denoted as ΔX, ΔY, and ΔZ, respectively. ΔZ can absorb elevation system deviations caused by support settlement or errors in measurement benchmarks. These variables collectively constitute the coordinate system calibrable parameter data. Mathematically, this can be represented as a part of a state vector containing geometric degrees of freedom. For example, the first four components of the state vector can be defined as [Δθ, ΔX, ΔY, ΔZ]. By setting these parameters as unknowns to be solved, the model possesses the ability to automatically align with the actual structure based on measured data in subsequent steps, rather than forcibly assuming that the design coordinate system and the actual coordinate system are completely coincident.

[0036] The key shape control factors for constructing the prestressed design curve are marked as calibrable curve parameter data. The key shape control factors include at least the control point sag correction or end slope correction within the curve segment.

[0037] Specifically, the mathematical description of the prestressed C4 curve is parametrically extended. The original C4 curve is usually composed of piecewise parabolic or circular curves, and its shape is uniquely determined by design parameters such as the tangent slope k and the position of the segment points. To adapt to changes in the alignment during actual construction, this embodiment introduces a correction factor. For example, the vertical position correction at the control point can be defined, denoted as Δ. k For example, a typical four-segment C4 curve can be represented by the equation y = f(x, k1, k2, k3), where x is the position variable and k1, k2, and k3 are design coefficients. In this embodiment, these fixed shape coefficients are converted into variable parameters, for example, defining the actual coefficient k1' = k1 + Δ. k1 ;where Δ k1 These are correction values ​​Δ. k This constitutes calibrable curve parameter data, reflecting the actual morphological changes of the prestressed duct under its own weight, support deformation, or construction errors. Incorporating these parameters into the state vector allows the model to describe flexible, deformable theoretical curves.

[0038] By linking the calibrable parameter data of the coordinate system and the calibrable curve parameter data to the local plane coordinate system data and the initial beam bottom reference surface data, a calibrable control space model with both fixed geometric properties and parameterized adjustment properties is generated.

[0039] In this embodiment, the final model is constructed through mathematical encapsulation. The system integrates the geometric coordinate system, initial reference surface, and all calibrable parameters defined above into a unified mathematical object. Specifically, the mathematical object contains a parameter vector X, whose structure can be represented as X = [Δθ, ΔX, ΔY, ΔZ, Δ...]. k1 Δ k2 Δ k3 At the same time, the model contains functional relationships that map these parameters to the three-dimensional coordinates of any point.

[0040] Furthermore, to ensure the physical rationality of the subsequent inversion process, a priori constraint mechanism is introduced. In one optional implementation, before constructing the parameter identification and optimization objective function, historical experience constraints are also introduced, specifically:

[0041] Historical engineering alignment adjustment records for beam spans of the same type are obtained, and the empirical distribution range of coordinate system adjustment and curve shape adjustment is extracted to generate prior constraint data. The prior constraint data is added as boundary conditions or regularization terms to the parameter identification optimization objective function to constrain the solution space of the self-calibration parameter data and prevent the inversion results from having abnormal solutions that violate the common sense of engineering physics.

[0042] In other words, based on historical engineering alignment adjustment records of beam spans of the same type, the empirical distribution range of each parameter is extracted. For example, based on experience, it is known that the rotation angle deviation is usually extremely small, so the value range of Δθ can be set between -1 degree and +1 degree. The system generates prior constraint data and adds it to the model as boundary conditions. The model generated in this way not only has adaptive adjustment capabilities but is also effectively constrained by engineering experience, preventing overfitting or divergence in the mathematical solution process.

[0043] like Figure 3 As shown, according to one aspect of this application, multi-point observation data is collected, including:

[0044] A parameter sensitivity analysis was performed on the calibrable controllable space model to calculate the degree of response of the theoretical sag at each position along the beam length to changes in the calibrable parameters of the coordinate system and the calibrable parameters of the prestress curve shape, and to obtain sensitivity distribution data.

[0045] In this embodiment, sensitivity analysis quantifies the impact of changes in each parameter on the final calculated sagitta. Specifically, the system defines the theoretical sagitta H as a function of the beam length position L and the parameter vector X, denoted as H = f(L, X). Sensitivity is specifically expressed as the partial derivative of this function with respect to each component of the parameter vector X. For the i-th position and the j-th parameter, its sensitivity coefficient S... ij =ΨH i / ΨX j , where Ψ is the partial derivative. In practical calculations, the finite difference method is preferred to approximate the solution of this partial derivative. For example, for a certain parameter X... j A small disturbance Δ is applied, and the change in sag before and after the disturbance is calculated. The ratio of the two changes is the approximate sensitivity value. The system iterates along the beam length, for example, every 0.1 meters, calculating the sensitivity of all positions to all parameters, forming a sensitivity matrix or curve graph. This visually shows which areas on the beam are most sensitive to the rotation angle and which areas are most sensitive to the vertical deformation of the curve, thus forming sensitivity distribution data.

[0046] Based on sensitivity distribution data, locations with a response to parameter changes higher than a preset threshold are selected as self-calibration measurement points, generating extended measurement point planning data that includes the location information of the self-calibration measurement points.

[0047] Specifically, sensitivity data is used for measurement point selection. According to information theory principles, measuring at locations with high sensitivity most effectively constrains the solution for the corresponding parameter. For example, the system sets optimization rules, such as finding the location L with the largest absolute value of sensitivity for each parameter to be identified. max For example, the analysis might reveal that the positions at both ends of the beam are most sensitive to the coordinate system rotation angle Δθ, while the mid-span position is most sensitive to the overall curve settlement ΔZ or shape factor Δ. k The beam end, mid-span, and peak position of the sensitivity curve are the most sensitive. Therefore, the system marks the beam end, mid-span, and peak position of the sensitivity curve as mandatory measurement points. Preferably, to ensure comprehensive coverage, zero-crossing points with alternating positive and negative sensitivity can also be selected as auxiliary measurement points. These selected key location coordinates and their corresponding measurement priorities are integrated to generate extended measurement point planning data, indicating where on-site surveyors should measure, thus avoiding the inefficiency and information redundancy caused by blindly and evenly distributing points.

[0048] Based on the extended measurement point planning data, the measurement equipment at the construction site is driven to capture the target, obtain the spatial coordinates of the self-calibrated measurement points, and form multi-point observation data for driving parameter inversion.

[0049] In this embodiment, digital planning is translated into physical actions. After receiving the planning data, surveyors or fully automated surveying robots move to the designated location to perform the operation. The equipment can utilize automatic target recognition technology to capture the prism center and obtain high-precision three-dimensional coordinate data. Furthermore, to ensure data quality, the acquisition process may include multiple repeated observations to eliminate random noise. The acquired raw data includes the coordinates of the station points, the coordinates of the backsight points, and the slope distance, horizontal angle, and zenith distance of each target measurement point, or directly outputs three-dimensional rectangular coordinates. All this data is aggregated and labeled with location information to form multi-point observation data. Because this data is acquired based on sensitivity planning, it contains the key geometric information required for analyzing model parameters, providing solid data support for subsequent high-precision parameter inversion.

[0050] This embodiment calculates the sensitivity of model parameters to observations, scientifically plans the location of measurement points, and obtains the most information-rich observation data with the fewest measurement points.

[0051] like Figure 4 As shown, in one possible implementation, the updated self-calibration control space model is generated, including:

[0052] Based on a calibrable controllable space model, the theoretical sag data of the corresponding positions of multi-point observation data under parameterized state are calculated, and the measured sag data of multi-point observation data are extracted.

[0053] In this embodiment, a mapping relationship is established between the observation space and the parameter space. Specifically, for each observation point i, the system determines its position x in the local coordinate system. i The theoretical height H is calculated using the current parametric C4 curve equation. Let the parametric curve equation be y = f(x, P), where P is a state vector containing coordinate parameters and a shape factor. theory_i =f(x i , P), where f is the function mapping of the curve equation. For the extraction of measured elevation data, the system needs to perform physical geometric conversions. Specifically, based on the prism center elevation Z measured by the total station... prism Based on auxiliary measurement parameters, the actual elevation of the beam bottom or corrugated pipe bottom is calculated. The calculation formula can be expressed as:

[0054] H actual_i = Z prism - H rod - T slab + H beam ;

[0055] Where H actual_i Z represents the measured sagitta of the corrugated pipe bottom or prestressed tendon bottom obtained from the conversion at the i-th measuring point; prismH represents the absolute elevation of the prism center obtained by total station measurement; rod Indicates the height of the prism rod used during measurement; T slab Indicates the thickness of the floor slab; H beam The beam height is represented by the value. In some embodiments, if the measurement point is located directly on the corrugated pipe, the plate thickness and beam height terms in the above formula can be adjusted or omitted according to the actual measurement location. Through conversion, the photoelectric measurement values ​​are converted into structural physical quantities to obtain the measured sag data.

[0056] A parameter identification and optimization objective function is constructed, with the coordinate system calibrable parameters and the prestress curve shape calibrable parameters as unknowns, and the goal of minimizing the deviation between theoretical and measured sagittal data.

[0057] Specifically, to recover the true parameters from noisy observation data, a mathematical optimization model needs to be constructed. The system defines the residual r. i Let H be the theoretical vector height of the i-th measurement point. theory_i Compared with the measured sagittal height H actual_i The difference. The parameter identification optimization objective function J(P) can be constructed as the sum of squares of the residuals at all measurement points, i.e., J(P) = ∑ i=1 n (r i 2 ), where n is the total number of observation points.

[0058] In a preferred implementation, to improve the algorithm's resistance to gross errors, a parameter identification optimization objective function is constructed, including:

[0059] Calculate the single-point residual data between the theoretical sag data and the measured sag data at each measuring point location.

[0060] Specifically, the system can perform preliminary fitting or use median estimation to obtain the initial residuals at each point.

[0061] Based on the magnitude of the single-point residual data, a corresponding residual weight coefficient is assigned to each measurement point; among them, measurement points whose single-point residual data exceed a preset threshold are assigned a lower weight value to reduce the impact of abnormal observations on the inversion results.

[0062] For example, Huber weighting function or IGG weighting function can be used. When the absolute value of the residual at a certain measurement point is less than a preset threshold k, its weight w... i Set to 1; when the absolute value of the residual is greater than k, the weight w i This can be set to k divided by the absolute value of the residual. This allows bad points that deviate too far from the normal trend to be automatically downweighted or even eliminated.

[0063] Based on the residual weight coefficient data, a weighted least squares form of weighted parameter identification optimization objective function is constructed, and robust self-calibration parameter data with noise resistance is obtained by solving the function.

[0064] In this embodiment, the objective function is changed to J(P)=∑ i=1 n (w i ·r i 2 This ensures that even if abnormal data is mixed in during field measurements due to prism rod misalignment or alignment deviations, the final calculated model parameters remain reliable.

[0065] Solve the objective function for parameter identification optimization to obtain the optimal parameter values ​​that minimize the deviation, and record the optimal parameter values ​​as self-calibration parameter data.

[0066] Preferably, a numerical optimization algorithm is used to solve the parameter identification optimization objective function; that is, a computer algorithm is used to find the parameter solution that minimizes the objective function. Since the C4 curve equation and coordinate transformation typically contain nonlinear terms, this problem belongs to the category of nonlinear least squares problems. The system can be solved iteratively using the Gauss-Newton method or the Levenberg-Marquardt algorithm.

[0067] In a specific numerical case, assume the state vector P to be solved contains the coordinate system rotation angle α and the Z-axis offset ΔZ. Initially, α is 0 and ΔZ is 0. Five measurement points were collected on-site, with initial calculated deviations of [12mm, 15mm, 18mm, 14mm, 10mm]. It can be seen that the deviations are generally large and exhibit a systematic trend. After iterative algorithm solving, the optimal parameter values ​​were identified as α = 0.05 radians and ΔZ = -10mm. Substituting these parameters into the model, the calculated residuals decreased to [2mm, -1mm, 2mm, -2mm, 0mm]. This indicates that the systematic error has been absorbed by the parameters, and only random error remains. This optimal solution is recorded as the self-calibration parameter data.

[0068] The coordinate system and curve shape in the calibrable control space model are corrected using self-calibration parameter data, generating a self-calibration control space model that includes the corrected coordinate system and the corrected prestress curve.

[0069] In this embodiment, the solved mathematical results are transformed into an engineering model. The system updates the coordinate transformation matrix and curve equation coefficients within the model. For example, the rotation matrix of the local coordinate system is updated to include the calculated rotation angle α, and the shape parameter k value of the C4 curve is updated to k+Δ. kThe self-calibrating control space model generated at this point represents the current digital twin state of the beam, and all subsequent measurements and layouts will be based on this model.

[0070] In a further embodiment, calculating the self-calibration elevation deviation data includes:

[0071] By using the corrected coordinate system in the self-calibrated control space model, the spatial coordinates in the multi-point observation data are remapped to the local space, and the self-calibrated measured sagittal data are calculated.

[0072] In this embodiment, the observation data is backtracked and remapped. The original observation data is absolute in the engineering coordinate system, but the relative positions in the local coordinate system change with the coordinate system correction. The system uses the updated coordinate transformation parameters to recalculate the local coordinates (x, y, y) of each measuring point relative to the new coordinate axes. new y new , z new The sag was then recalculated based on this data to obtain the self-calibrated measured sag data. This ensured the consistency of the evaluation benchmark.

[0073] Using the corrected prestress curve in the self-calibration control space model, the self-calibration design sagitta data corresponding to each measuring point position are calculated.

[0074] Specifically, using the C4 curve equation containing correction parameters, the new local abscissa x of each measuring point is substituted. new The theoretical height that should be reached at that position is calculated, which is the self-calibrated design sag data, representing the ideal target after adapting to structural deformation.

[0075] The difference between the self-calibrated measured sag data and the self-calibrated design sag data is calculated to generate self-calibrated sag deviation data that reflects the degree of deviation of the construction alignment from the adaptive target alignment.

[0076] In this embodiment, the final quality evaluation index is output. The final deviation is obtained by subtracting the measured value from the design value. This deviation eliminates the influence of coordinate system establishment errors and systematic structural deformation, purely reflecting the local quality of the corrugated pipe installation. In some optional implementations, the system can also calculate the root mean square error of this deviation data as a comprehensive index for evaluating the overall construction quality of the beam.

[0077] This embodiment constructs a numerical optimization problem, uses the weighted least squares method to process the observed data, calculates the true structural parameters, thereby eliminating systematic errors, and recalculates high-precision construction deviations based on the corrected model.

[0078] Obtain engineering tolerance specifications and measurement equipment technical indicators corresponding to standardized design parameter data, extract statistical distribution characteristics of plate thickness deviation, beam height deviation and measurement noise, and generate uncertain prior data.

[0079] In this embodiment, the system quantifies the uncertainty of the input data. Specifically, it consults relevant engineering construction quality acceptance specifications to obtain the allowable deviation ranges for slab thickness and beam height; for example, the allowable deviation for slab thickness is ±5mm. Based on statistical principles, it can be assumed that these structural dimensions follow a normal distribution, and their allowable deviation ranges are converted into standard deviations. For example, the standard deviation σ for slab thickness is set. t This equals the allowable deviation divided by 3. Simultaneously, based on the technical specifications of the measuring equipment, such as a total station, obtain its distance measurement accuracy and angle measurement accuracy, and calculate the standard deviation σ of the elevation measurement. noise These statistical parameters collectively constitute the uncertain prior data. In some alternative implementations, the statistical distribution of measured data accumulated from historical projects can be used to replace the standard values ​​to obtain prior data that better reflects the skill level of a specific construction team.

[0080] Establish a mapping relationship between uncertain prior data and standardized design parameter data, assign corresponding standard deviation or confidence interval attributes to the geometric dimensions in the calibrable control space model, and generate uncertain structural dimension data.

[0081] Specifically, statistical attributes are linked to specific physical parameters. The system adds a variance or standard deviation field to each key dimensional parameter in the data structure. For example, the plate thickness parameter is represented as T. slab =T nominal +N(0,σ) t 2 ), where T nominal Let N be the nominal or design value of the plate thickness, and let N() be a normally distributed random variable. This dataset with probability distribution properties is the uncertain structural dimension data, which provides an input source for subsequent error propagation analysis.

[0082] Uncertain structural size data are encapsulated into a calibrable control space model to generate an uncertainty-enhanced control space model with error statistical properties.

[0083] In this embodiment, the model's attributes are upgraded. The uncertainty-enhanced control space model not only includes geometric equations for calculating nominal values ​​but also integrates a covariance matrix for calculating variance. Specifically, the model internally maintains a parameter covariance matrix ∑ P Its diagonal elements correspond to the variance of each input parameter, and its off-diagonal elements correspond to the correlation between parameters.

[0084] In a further embodiment, a probabilistic risk assessment of construction quality is also included, specifically:

[0085] Based on the error statistical properties defined by the uncertainty-enhanced control space model, and combined with the geometric transfer path determined by the self-calibration control space model, the variance or confidence interval of the sag deviation at each position along the beam length direction is calculated using a pre-configured error propagation model, generating sag uncertainty distribution data.

[0086] In this embodiment, the variance of the output can be calculated using the error propagation law. The system calculates the sensitivity matrix J of the height calculation function with respect to each input parameter; this matrix is ​​composed of partial derivatives. Based on the linear error propagation formula, the variance Var of the output height is calculated. H Var H =J·∑ P ·J T ,in T For transpose; Var H ∑ represents the variance of the calculated elevation deviation at a certain position; P This represents the covariance matrix of input parameters (such as plate thickness and beam height), with the diagonal elements being the variances of each parameter. Preferably, the influence of measurement noise can also be superimposed, i.e., the total variance σ. total 2 =Var H +σ noise 2 , where σ noise This represents the standard deviation of random noise from the measuring equipment itself. The system performs this calculation point-by-point along the beam length to obtain the standard deviation of the sag at each location. Based on this, confidence intervals for the sag can be constructed; for example, a 95% confidence interval is [H]. mean -1.96·σ total H mean +1.96·σ total ], where H mean σ is the expected or mean value of the theoretical height. total This represents the total standard deviation. This series of statistics, distributed with respect to location, constitutes the data with sag uncertainty distribution.

[0087] By introducing a preset engineering allowable deviation threshold, the probability value of the sag deviation exceeding the limit at each measuring point is calculated based on the sag uncertainty distribution data, and risk assessment data indexed by location is generated.

[0088] Specifically, the statistical distribution is transformed into an intuitive risk probability. Let the maximum allowable deviation of the elevation in the project be Δ. limit For any position x, the deviation follows a mean of μ. x (Current measured deviation), standard deviation is σ total _ xThe system follows a normal distribution. The absolute value of the calculated deviation is greater than Δ. limit The probability P fail For example, this can be achieved using the cumulative distribution function (CDF) of the standard normal distribution:

[0089] P fail = 1 - (CDF((Δ limit - μ x ) / σ total_x ) - CDF((-Δ limit - μ x ) / σ total_x ));

[0090] Where P fail Δ represents the probability that the elevation deviation of a measuring point exceeds the allowable range; CDF represents the cumulative distribution function of the standard normal distribution; Δ limit This indicates the maximum allowable deviation threshold according to engineering specifications, such as 5 mm; μ x This represents the mean current elevation deviation at that location, i.e., the measured deviation; σ total_x This represents the total standard deviation of the sag at that position, i.e., Var H The square root of the value. The calculation results are organized into risk assessment data indexed by location. For example, the data structure can be represented as: {Location: L=15m, Current deviation: 4mm, Uncertainty: 2mm, Probability of exceeding limit: 15%}.

[0091] In a further embodiment, adaptive control is performed based on risk assessment data, specifically:

[0092] Retrieve risk assessment data, filter out beam segments with a probability of exceeding limits higher than the preset safety level, and mark the filtered beam segments as a set of high-risk locations.

[0093] In this embodiment, risk classification and screening are performed. Specifically, the system sets a safety level threshold, for example, 10%. It iterates through all calculation points and assigns the probability of exceeding the limit P... fail Locations accounting for more than 10% of the total are identified and categorized into a high-risk location set. These locations are often areas with complex structural deformations, high sensitivity, or where the current deviation is approaching a critical value.

[0094] Control instructions are generated for high-risk location sets to increase the number of encrypted observations or repeated measurements, forming risk control point data to guide the construction site in carrying out enhanced measurements and adjustments in the corresponding risk areas.

[0095] Specifically, for each point in the high-risk set, the system generates a corresponding control strategy. For example, for points with an extremely high probability of exceeding limits, it generates an encrypted observation instruction, requiring the addition of auxiliary measurement points before and after that point; or it generates a re-measurement instruction, requiring at least three independent measurements of that point to reduce random noise σ. noise The impact of these instructions constitutes risk control point data, which, as specific action plans, are directly issued to surveying equipment or construction teams, achieving a closed loop from data analysis to on-site actions.

[0096] This embodiment extends deterministic geometric calculation to statistical inference of random variables, realizing risk-driven intelligent control.

[0097] In one possible embodiment, the method further includes performing on-site stakeout control based on the self-calibration results, specifically:

[0098] During the bellows positioning or acceptance phase, the self-calibrating control space model is invoked in real time to calculate the target layout coordinate data of the current layout position under the corrected target line shape, and drive the fully automatic measuring equipment to indicate the target position.

[0099] In this embodiment, the digital model is transformed into physical guidance. When construction workers need to locate a specific corrugated pipe, they input the mileage value or select the point to be laid out via a handheld terminal. The system backend then calls the latest self-calibrating control space model in real time and calculates the theoretical coordinates (x, y, x) of the point in the local coordinate system based on the corrected curve equation. target y target , z target Using the corrected coordinate transformation parameters, the local coordinates are inversely transformed into absolute coordinates (X, Y) in the engineering survey coordinate system. global Y global Z global The system transmits the absolute coordinates to a fully automatic total station via a wireless communication interface. Upon receiving the command, the total station automatically rotates the alidade and telescope, emitting a laser to indicate the physical location of the three-dimensional coordinates in space, or automatically tracks the prism and guides the worker to the target location. This achieves intelligent lofting based on adaptive target alignment.

[0100] The self-calibrated sag deviation data is displayed in real time on the display interface. When the deviation data exceeds the preset range, a construction adjustment instruction is generated to guide the construction personnel to adjust the height of the corrugated pipe support until the deviation meets the requirements.

[0101] Specifically, the handheld terminal's display interface is divided into a graphics area and a data area. The graphics area displays the design alignment of the prestressed curve and the relative position of the current measured point; the data area displays the self-calibrated sag deviation data in real time. Preferably, to improve on-site work efficiency, the system adopts a color coding mechanism: when the deviation is less than a preset acceptable threshold, such as 5mm, the value is displayed in green and accompanied by a qualified prompt sound; when the deviation is greater than the threshold, the value is displayed in red and accompanied by an alarm sound. Simultaneously, the system generates specific text or voice construction adjustment instructions based on the sign and magnitude of the deviation, such as "The support is 12mm too low, please adjust it upwards." After the worker adjusts the corrugated pipe support according to the instructions, the system drives the total station to measure again, refreshing the deviation value in real time. This process is repeated until the deviation value turns green, ensuring that the construction quality of each control point strictly meets the standards. This provides intuitive visual feedback and closed-loop control.

[0102] This embodiment constructs a complete closed-loop technical system from parametric modeling, sensitivity planning, robust self-calibration inversion, uncertainty risk assessment to intelligent visualization layout, effectively solving the problem of insufficient construction control accuracy caused by model solidification and neglect of uncertainty.

[0103] In one exemplary embodiment, constructing local planar coordinate system data includes:

[0104] The location information of the corner point at the intersection of the floor slab and the beam is extracted from the standardized design parameter data as the reference point for station construction, and the position of the prestressing start point along the layout direction is defined as the origin of the coordinate system. Based on the station construction reference point and the origin of the coordinate system, the direction from the prestressing start point to the prestressing end point is defined as the first axis of the local plane coordinate system, and the direction perpendicular to the first axis and parallel to the floor slab plane is defined as the second axis of the local plane coordinate system, thereby generating local plane coordinate system data that strictly corresponds to the physical position of the engineering entity.

[0105] In this embodiment, let P be the coordinates of the floor slab corner point in the engineering coordinate system (global coordinate system). corner (X c Y c Z c The coordinates of the prestressing start point are P. start (X s Y s Z s The coordinates of the prestressing endpoint are P. end (X e Y e Z e Calculate the unit vector u along the X-axis of the local coordinate system. x :

[0106] u x =(P end -Pstart ) / ∣∣P end -P start ||;

[0107] Where P end -P start Let |P| be the vector difference. end -P start || represents the magnitude of the vector difference. In most architectural engineering projects, the Z-axis is assumed to be perpendicular to the horizontal plane and pointing upwards, i.e., the unit vector u... z The value is (0, 0, 1). However, considering the potential design slope of the floor slab, it is preferable to use the normal vector of the floor slab plane as the local Z-axis direction. If the floor slab is assumed to be horizontal, then u z Take the coordinates (0, 0, 1). Calculate the local Y-axis unit vector u using the vector cross product. y :u y =u z ×u x At this point, construct the rotation matrix R: R=[u x u y u z ] T Establish from global coordinates P global To local coordinates P local Conversion formula:

[0108] P local = R * (P global - P start );

[0109] Where P local The coordinate vector (x) of the measuring point in the local plane coordinate system. local y local , z local R represents the rotation matrix from the global coordinate system to the local planar coordinate system, which consists of the three axial unit vectors u of the local coordinate system. x u y u z Composition; P global P represents the measured coordinate vector (X, Y, Z) of the measuring point in the global coordinate system of the project; start This represents the coordinate vector of the prestressing initiation point in the global coordinate system of the project, serving as the origin of the local coordinate system. Through linear algebraic transformation, the coordinates of any measuring point on site are mapped to a local space with the prestressing initiation point as the origin and the prestressing tendon orientation as the X-axis, providing a unified geometric benchmark for subsequent curve equation calculations.

[0110] Furthermore, key shape control factors are constructed for the prestressed design curve and marked as calibrable curve parameter data. These key shape control factors include at least the control point elevation correction or end slope correction within the curve segments. Specifically, the initial mathematical equation of the prestressed C4 curve in the local coordinate system is the basic form of the model before parameterization correction. The C4 curve is divided into four segments, and its design elevation y has the following functional relationship with the local abscissa x:

[0111] First segment (0≤x≤K1·L): ;

[0112] The second segment (K1·L≤x≤K2·L): ;

[0113] The third segment (K2·L≤x≤L-K3·L): ;

[0114] The fourth paragraph (L-K3·L≤x≤L): ;

[0115] Where y represents the design sag of the prestressing tendon in the local coordinate system, i.e., the vertical height from the bottom of the beam or the reference plane; x represents the lateral position of the prestressing tendon in the local coordinate system, i.e., the horizontal distance from the prestressing start point; L is the beam span length; h is the beam height; e1, e2, and e3 are the protective layer thickness or control distance of the prestressing tendon from the bottom or top of the beam at the start, mid-span, and end points, respectively; k1, k2, and k3 are dimensionless curve segmentation scaling coefficients. In the parametric model of this embodiment, the parameters h, L, e1, e2, e3, k1, k2, and k3 in the above formulas are no longer considered absolute constants, but are superimposed with correction terms to be identified, such as h actual =h design +Δ h This transforms the static equation into a dynamic, calibrable curve model, where h actual h is the actual value. design Δ is the nominal or theoretical value determined during the design phase. h This is a correction amount.

[0116] According to one aspect of this application, constructing calibrable controllable spatial model data can further involve: reading design parameter data from design drawings and structural calculation sheets, wherein the design parameter data includes beam span geometry information, prestressing curve type information, prestressing design elevation information, and control section location annotation information; and simultaneously extracting structural dimension data from structural construction drawings, wherein the structural dimension data includes beam height, slab thickness, relative positional relationship between beam ends and construction site, and elevation benchmark information of relevant components. The design parameter data and structural dimension data are then converted into a unified engineering coordinate system expression, and the unit system and coordinate direction conventions are verified to obtain standardized design parameter data and standardized structural dimension data with consistent format that can be used for subsequent modeling.

[0117] Based on standardized structural dimension data, the corner point at the junction of the floor slab and beam is selected as a local construction site. The coordinates of this construction site in the engineering coordinate system and the relative positional relationship between the prestressing initiation point and the construction site are read. The prestressing initiation point along the prestressing tendon placement direction is defined as the origin of the local plane coordinate system, the prestressing endpoint direction is defined as the first axis of the local plane coordinate system, and the direction perpendicular to the prestressing endpoint direction and parallel to the floor slab plane is defined as the second axis of the local plane coordinate system. This yields local plane coordinate system data describing the beam-span plane positional relationship. Pre-defined coordinate translation and rotation relationships can be used to convert the coordinates of any plane point in the engineering coordinate system into the expression form in the local plane coordinate system data, ensuring that subsequent prestressing curves and measurement points can be uniformly described within the same local plane system. Furthermore, while constructing the local plane coordinate system data, adjustable parameters for self-calibration can be introduced. The plane rotation angle and origin translation offset of the local plane coordinate system are recorded as calibrable parameters, resulting in calibrable local plane coordinate system data including both fixed and adjustable parts.

[0118] Based on calibrable local plane coordinate system data and standardized structural dimension data, the vertical coordinates of the prestressing initiation beam bottom position relative to the local plane coordinate system are determined according to the geometric relationship between the design slab thickness, design beam height, and construction site elevation. This position is defined as the vertical zero point of the local spatial coordinate system, constructing local spatial coordinate system data containing three dimensions: plane and vertical. The space containing the beam bottom reference plane, the upper surface of the floor slab, and the design alignment of the prestressing duct is uniformly expressed under the local spatial coordinate system data, obtaining the initial beam bottom reference plane data for subsequent prestressing sag calculation. Furthermore, to support uncertainty analysis, while constructing the local spatial coordinate system data and the initial beam bottom reference plane data, a mapping relationship is established between the uncertain prior data and structural dimensions such as slab thickness, beam height, and construction site elevation. Each structural dimension is assigned a corresponding standard deviation or interval range, forming uncertain structural dimension data with attached statistical attributes. On this basis, the uncertain structural dimension data is linked to the local spatial coordinate system data and the initial beam bottom reference plane data, forming uncertainty-enhanced local spatial coordinate system data with geometric coordinate attributes and uncertainty attributes.

[0119] Based on standardized design parameter data, when generating initial curve model data, key shape parameters of the curve, such as the sag of control points within segments, end slopes, and mid-span sag corrections, are recorded as adjustable parameters, forming calibrable curve parameter data with a set of shape-adjustable parameters. Simultaneously, this calibrable curve parameter data is combined with calibrable local plane coordinate system data and initial beam bottom reference surface data to generate calibrable control space model data with calibrable characteristics. In some preferred embodiments, prior constraint data can also be introduced into the calibrable curve parameter data to set reasonable value ranges and initial estimates for each shape-adjustable parameter. During subsequent self-calibration, constraints prevent abnormal curve shapes that do not conform to engineering experience, improving the stability and convergence of the self-calibration solution.

[0120] Simultaneously, calibrable local plane coordinate system data, uncertainty-enhanced local space coordinate system data, and calibrable control space model data are uniformly encapsulated to form enhanced control space model data containing fixed parameters, adjustable parameters, and uncertainty attributes. The model version number, parameter initial state, and uncertainty configuration are recorded in this data for use when executing self-calibration algorithms and risk assessments.

[0121] According to another aspect of this application, the calculation of self-calibration sag deviation data can also be as follows: Based on self-calibration measurement observation data and enhanced control space model data, using the coordinate transformation relationship in the control space model, the engineering coordinates and instrument elevations of the measurement points in the measurement observation data are converted into measurement point coordinates under the local space coordinate system data; using the vertical position of the beam bottom reference plane in the local space coordinate system, the difference between the elevation of the measurement point and the elevation of the beam bottom reference plane is calculated as the measured sag data of that measurement point. The measured sag data of all measurement points are organized according to the corresponding beam length position index for subsequent point-by-point comparison with the design sag. Furthermore, to facilitate subsequent uncertainty analysis, while calculating the measured sag data, based on the uncertainty structural dimension information in the uncertainty-enhanced local space coordinate system data, an error propagation coefficient caused by both structural dimensions and measurement noise is added to each measured sag, generating enhanced measured sag data containing the mean and error coefficient.

[0122] Based on enhanced measured sag data and enhanced control space model data, at each measurement point along the beam length, the design sag data for that location is calculated using the initial curve model. The measured sag data is then subtracted from the design sag data point by point to obtain initial sag deviation data reflecting the degree of construction deviation. This initial sag deviation data is sorted and summarized by measurement point number or beam length position to guide construction adjustments and acceptance evaluation. Furthermore, the sensitivity coefficient of the design sag to curve parameters is recorded for each location, generating enhanced design sag data containing design values ​​and sensitivity information. This data is used to construct the objective function and Jacobian matrix for self-calibration parameter identification, thereby improving solution efficiency and stability.

[0123] Based on self-calibrated measurement observation data, calibrable control space model data, enhanced design elevation data, and enhanced measured elevation data, a parameter identification model is constructed with calibrable parameters of the coordinate system and calibrable parameters of the curve shape as unknowns. The difference between the measured elevation at each calibration point and the theoretical elevation calculated by the calibrable control space model is used as the residual term. An optimization problem with the sum of squared residuals as the objective is established, and a set of optimal self-calibrated parameter data is obtained by solving it. The system uses the self-calibrated parameter data to synchronously update the calibrable local plane coordinate system data and the calibrable curve parameter data, generating self-calibrated control space model data containing the corrected coordinate system and the corrected prestressed curve, which is used for subsequent recalculation of elevation deviations and construction control. Furthermore, to improve the robustness of the self-calibration results to anomaly observations, different weights are assigned to the residual terms of each calibration measurement point when constructing the parameter identification model. The weights of measurement points with larger residuals are reduced, and an optimization objective of weighted least squares or robust regression is constructed to obtain robust self-calibration parameter data that is insensitive to anomalies. Based on these robust parameters, the self-calibration control space model data is updated to enhance the applicability of the model in complex construction environments.

[0124] Based on the self-calibrated control space model data and self-calibrated measurement observation data, coordinate transformation and sag calculation are re-performed. The spatial coordinates of all measurement points are transformed in the self-calibrated coordinate system to calculate the self-calibrated measured sag data. The self-calibrated design sag data for the corresponding locations is then calculated using the self-calibrated prestressing curve. Subtracting the two yields the self-calibrated sag deviation data, reflecting the construction quality under the corrected target alignment. The system compares the initial sag deviation data with the self-calibrated sag deviation data to evaluate the improvement effect of the self-calibration process on the overall deviation distribution, providing a basis for subsequent construction adjustment strategies.

[0125] Based on uncertain prior data, uncertainty-enhanced control space model data, enhanced measured sag data, and self-calibrated control space model data, a sag error propagation model is constructed. Using linear error propagation formulas or numerical simulation methods, the variance or confidence interval of the sag values ​​at each location within the beam length is calculated, generating sag uncertainty distribution data reflecting the uncertainty level of sag variation with location. On this basis, combined with engineering allowable deviation thresholds, the probability of sag exceeding limits at each location is calculated as risk assessment data. Based on the magnitude of the risk probability, a set of locations requiring key control is selected, forming risk control point data for subsequent layout and acceptance. In some preferred implementations, to achieve comprehensive optimization of risk and cost, the construction costs of adding measuring points, adding supports, or increasing the number of remeasurements can be considered simultaneously when constructing the risk control point set. The risk of exceeding limits and control costs are jointly incorporated into the evaluation function. Heuristic algorithms are used to generate optimized risk control scheme data with the lowest risk under cost constraints, guiding layout and priority adjustment.

[0126] In summary, the intelligent measurement method for prestressed sag height includes: constructing a calibrable control space model containing coordinate system parameters and curve shape parameters; collecting multi-point observation data on site, inverting and solving the model parameters by minimizing the objective function of the theoretical and measured sag height residuals, and generating a self-calibrating control space model that fits the actual physical state; using an error propagation model to assess the risk of sag height exceeding the limit, and optimizing the layout of measurement points based on the risk distribution.

[0127] This invention employs parametric modeling and multi-point inversion self-calibration. By constructing a calibrable model incorporating rotation, translation, and shape factors, and utilizing multi-point field observation data to build a least-squares optimization problem, it can accurately identify and calculate the actual deflection of the coordinate system and the actual deformation of the curve. Systematic structural deviations, such as support settlement and overall rotation, are separated from the total deviation and the evaluation benchmark is updated. This ensures that the final output self-calibrated sag deviation only reflects the local installation quality of the bellows, solving the problem of misjudgment caused by benchmark errors. Uncertainty modeling and risk adaptive control are introduced. By modeling plate thickness, beam height, and equipment noise as random variables and using the Jacobian matrix for linear error propagation calculation, the sag uncertainty distribution along the beam length is quantified, and risk assessment data is generated accordingly. Based on this risk assessment, the system can automatically generate encrypted observation instructions in areas with high probability of exceeding limits, solving the problem that traditional uniform point distribution strategies may miss key defects in error-sensitive areas such as inflection points and mid-span, achieving scientific and refined quality control.

[0128] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for intelligent measurement of prestressed height, characterized in that, The method comprises the following steps: Based on the obtained standardized design parameter data, a calibratable control space model is constructed, including preset coordinate system calibratable parameters and prestressed curve shape calibratable parameters; Multi-point observation data is collected at predetermined positions on the construction site, including the spatial coordinate information of each measuring point; The multi-point observation data is combined with the calibratable control space model, and the coordinate system calibratable parameters and the prestressed curve shape calibratable parameters are inversely solved by minimizing the residual error between the theoretical height and the measured height, to generate an updated self-calibration control space model; Based on the self-calibration control space model, self-calibration height deviation data corresponding to the multi-point observation data is calculated.

2. The method of claim 1, wherein, The method for constructing the calibratable control space model comprises the following steps: Based on the standardized design parameter data, the beam span geometric information and the structural size information are analyzed, and the local plane coordinate system data and the initial beam bottom reference surface data describing the basic geometric relationship of the beam body are constructed; The plane rotation angle and the original point translation offset of the local plane coordinate system data relative to the preset engineering survey coordinate system are constructed, which are marked as coordinate system calibratable parameter data; The key shape control factors of the prestressed design curve are constructed, which are marked as calibratable curve parameter data, and the key shape control factors at least include the control point height correction or the end slope correction in the curve segment; The coordinate system calibratable parameter data and the calibratable curve parameter data are connected to the local plane coordinate system data and the initial beam bottom reference surface data, to generate a calibratable control space model with fixed geometric properties and parameterized adjustment properties.

3. The method of claim 1, wherein, The method for collecting multi-point observation data comprises the following steps: Performing parameter sensitivity analysis on the calibratable control space model, calculating the response degree of the theoretical height at each position along the beam length to the change of the coordinate system calibratable parameters and the prestressed curve shape calibratable parameters, and obtaining sensitivity distribution data; Based on the sensitivity distribution data, selecting positions with a response degree higher than a preset threshold to parameter change as self-calibration measuring points, and generating extended measuring point planning data containing the position information of the self-calibration measuring points; According to the extended measuring point planning data, driving the measuring equipment to capture targets on the construction site, obtaining the spatial coordinates of the self-calibration measuring points, and composing multi-point observation data for driving parameter inversion.

4. The method of claim 1, wherein, The method for generating the updated self-calibration control space model comprises the following steps: Based on the calibratable control space model, calculating the theoretical height data of the positions corresponding to the multi-point observation data in the parameterized state, and extracting the measured height data in the multi-point observation data; Constructing a parameter identification optimization objective function taking the coordinate system calibratable parameters and the prestressed curve shape calibratable parameters as unknown quantities and minimizing the deviation between the theoretical and measured height data as the target; Solving the parameter identification optimization objective function to obtain the optimal parameter value that minimizes the deviation, and recording it as self-calibration parameter data; Using the self-calibration parameter data to correct the coordinate system and the curve shape in the calibratable control space model, to generate a self-calibration control space model containing the corrected coordinate system and the corrected prestressed curve.

5. The method of claim 4, wherein, The method for constructing the parameter identification optimization objective function comprises the following steps: Calculating the single-point residual error data between the theoretical height data and the measured height data of each measuring point position; According to the numerical size of the single-point residual data, a corresponding residual weight coefficient data is assigned to each measuring point; wherein, a lower weight value is given to the measuring point whose single-point residual data exceeds the preset threshold value; Based on the residual weight coefficient data, a weighted parameter identification optimization objective function in the form of weighted least squares is constructed, and the robust self-calibration parameter data with noise resistance is obtained by solving.

6. The method of claim 4, wherein, The self-calibration height deviation data is calculated, including: Using the modified coordinate system in the self-calibration control space model, the spatial coordinates in the multi-point observation data are remapped to the local space, and the self-calibration measured height data is calculated; Using the modified prestressed curve in the self-calibration control space model, the self-calibration design height data corresponding to the position of each measuring point is calculated; The difference between the self-calibration measured height data and the self-calibration design height data is calculated, and the self-calibration height deviation data reflecting the deviation degree of the construction alignment from the adaptive target alignment is generated.

7. The method of claim 1, wherein, It also includes constructing a statistical model for risk analysis, specifically: Obtain the engineering tolerance specification corresponding to the standardized design parameter data and the measurement equipment technical index information, extract the statistical distribution characteristics of the plate thickness deviation, beam height deviation and measurement noise, and generate the uncertainty prior data; Establish the mapping relationship between the uncertainty prior data and the standardized design parameter data, give the corresponding standard deviation or confidence interval attribute to the geometric size in the calibrable control space model, and generate the uncertainty structural size data; Encapsulate the uncertainty structural size data into the calibrable control space model to generate the uncertainty enhanced control space model with error statistical properties.

8. The method of claim 7, wherein, Further including probability risk assessment of construction quality, specifically: Based on the error statistical properties defined by the uncertainty enhanced control space model, combined with the geometric transmission path determined by the self-calibration control space model, the variance or confidence interval of the height deviation at each position along the beam length direction is calculated using the pre-configured error propagation model, and the height uncertainty distribution data is generated; Introduce the preset engineering allowable deviation threshold, calculate the probability value of the height deviation exceeding the limit at each measuring point position based on the height uncertainty distribution data, and generate the risk assessment data indexed by position.

9. The method of claim 8, wherein, Further including adaptive control based on risk assessment data, specifically: Retrieve the risk assessment data, filter out the beam segment positions with exceeding probability higher than the preset safety level, and mark them as a high-risk position set; Generate control instructions for encrypted observation or increased number of repeated measurements for the high-risk position set, and form risk control point data.

10. The method of claim 4, wherein, Before constructing the parameter identification optimization objective function, it also includes introducing historical experience constraints, specifically: Obtain the historical engineering alignment adjustment records of the same type of beam span, extract the experience distribution range of the coordinate system adjustment amount and the curve shape adjustment amount, and generate the prior constraint data; Add the prior constraint data as boundary conditions or regularization terms to the parameter identification optimization objective function to constrain the solution space of the self-calibration parameter data.

Images