BIM-based digital pre-assembling method and system for glass curtain wall

Abstract

The invention relates to a BIM (Building Information Modeling)-based digital pre-assembly method and system for a glass curtain wall, and relates to the technical field of data processing, and the method comprises the following steps: determining geometric parameters and installation position parameters of each component; the method comprises the following steps: acquiring point cloud data of a construction site through three-dimensional laser scanning, performing multi-source data registration on the point cloud data, and calculating to obtain a point cloud registration error; determining a feature matching degree based on the point cloud registration error; obtaining a positioning optimization target according to a deviation analysis result and an engineering constraint condition; according to the positioning optimization target, an assembly precision evaluation value is obtained through calculation; based on the assembly precision evaluation value, the correction value of the glass curtain wall in the pre-assembly process is determined; and calculating a comprehensive quality score based on the assembly precision evaluation value, the point cloud registration error, the feature matching degree and the correction value, and determining to adopt a corresponding pre-assembly scheme when the comprehensive quality score is greater than a preset threshold value. The pre-assembly accuracy of the glass curtain wall can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a BIM-based digital pre-assembly method, system, electronic equipment, and non-transient computer-readable storage medium for glass curtain walls. Background Art

[0002] Glass curtain walls, a widely adopted exterior envelope structure in modern architecture, offer advantages such as aesthetics, transparency, and lightweight construction. They are widely used in high-rise buildings, commercial complexes, and landmark projects. Currently, curtain wall construction relies primarily on two-dimensional drawings and manual experience. During the pre-assembly phase, traditional manual measurement and positioning methods are often used. Pre-assembly accuracy is highly dependent on the construction workers' skills and site conditions.

[0003] However, two-dimensional drawings cannot accurately express the installation relationship of curtain wall components in space, resulting in frequent errors in the pre-assembly process and a high rework rate. On the other hand, traditional manual measurement and assembly methods cannot achieve real-time monitoring and correction of the position, posture and assembly accuracy of curtain wall components, affecting construction quality and efficiency. Summary of the Invention

[0004] In response to the technical problems existing in the prior art, the present invention provides a BIM-based digital pre-assembly method, system, electronic device and non-transitory computer-readable storage medium for glass curtain wall, which can improve the accuracy of glass curtain wall pre-assembly.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides a BIM-based digital pre-assembly method for a glass curtain wall, the method comprising:

[0007] Establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component;

[0008] Acquire point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error;

[0009] Determining a feature matching degree based on the point cloud registration error;

[0010] According to the deviation analysis results and engineering constraints, the positioning optimization target is obtained;

[0011] According to the positioning optimization goal, an assembly accuracy evaluation value is calculated;

[0012] Determining a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value;

[0013] A comprehensive quality score is calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount. When the comprehensive quality score is greater than a preset threshold, the corresponding pre-assembly solution is determined to be adopted.

[0014] Optionally, performing multi-source data registration on the point cloud data to calculate a point cloud registration error includes:

[0015] Each BIM model point is transformed according to the rotation matrix and translation vector to fit each corresponding real-world cloud point to obtain the main registration item;

[0016] Processing the rotation transformation gradient of the rotation matrix and the translation transformation gradient of the translation vector according to the regularization parameter to obtain a regularization term;

[0017] A point cloud registration error is calculated based on the main registration term and the regularization term.

[0018] Optionally, the point cloud registration error is expressed as:

[0019]

[0020] Where E is the point cloud registration error, w i is the weight coefficient, p i and q i is the point pair matched between BIM and point cloud, p i is the BIM model point, q i is the real-world point cloud, λ is the regularization parameter, R is the rotation matrix, t is the translation vector, is the rotation transformation gradient, is the translation transformation gradient.

[0021] Optionally, determining the feature matching degree based on the point cloud registration error includes:

[0022] Obtain the sum of the distances between the BIM model point cloud and the actual point cloud after iterative closest point registration algorithm.

[0023] Obtaining a sum of feature similarities between each of the structures and the corresponding real-scene component in terms of local geometric features, surface normal, reflection intensity, color, and shape;

[0024] Obtaining a sum of topological consistencies indicating consistency in relative positions and connection relationships between each of the structures and the corresponding real-scene components;

[0025] The feature matching degree is obtained by performing a weighted summation on the sum of the point pair distances, the sum of the feature similarities, and the sum of the topological consistency.

[0026] Optionally, the feature matching degree is expressed as:

[0027] M=α∑ICP+β∑F+γ∑D;

[0028] Among them, M is the feature matching degree, ICP is the iterative closest point distance, F is the feature similarity, D is the topological consistency, α, β, and γ are the first weight, second weight, and third weight, respectively.

[0029] Optionally, obtaining a positioning optimization target based on the deviation analysis results and engineering constraints includes:

[0030] Determining a spatial offset error based on an initial position of the point cloud measurement and a target position corresponding to the initial position;

[0031] Determine the attitude deviation based on the attitude angle measured by the point cloud and the target attitude angle corresponding to the attitude angle;

[0032] Determining an error in the seam size based on a seam size measured by the point cloud and a target seam size corresponding to the seam size;

[0033] The spatial offset error, the posture deviation, and the error of the seam size are weightedly summed to obtain the positioning optimization target.

[0034] Optionally, calculating the assembly accuracy evaluation value according to the positioning optimization target includes:

[0035] Determining the geometric measuring point deviation of each key measuring point according to the measuring point deviation, the corresponding deviation weight and the total number of key measuring points;

[0036] Determine the joint error exponential function according to the correction coefficient, the joint error and the total number of joint measurement points;

[0037] An assembly accuracy evaluation value is calculated based on the geometric measurement point deviation and the joint error exponential function.

[0038] Optionally, determining a correction amount in a pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value includes:

[0039] Obtaining a current position deviation representing a currently measured spatial position offset between the component and the target position;

[0040] Calculate the derivative of the accumulated error and determine the error change rate compensation;

[0041] Calculating the integral of the accumulated error to determine an integral correction term;

[0042] The correction amount is determined according to the current position deviation, the error change rate compensation and the integral correction term.

[0043] Optionally, the comprehensive quality score calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree, and correction amount includes:

[0044] Determine the assembly accuracy contribution item according to the assembly accuracy evaluation value;

[0045] Determining a point cloud error term according to the point cloud registration error and a maximum value of the point cloud registration error;

[0046] Determining a feature matching item according to the feature matching degree;

[0047] Determine the correction capability item based on the correction amount and the maximum acceptable correction threshold;

[0048] The comprehensive quality score is determined according to the assembly accuracy contribution item, the point cloud error item, the feature matching item, and the correction capability item.

[0049] The present invention also provides a BIM-based digital pre-assembly system for glass curtain walls, the system comprising:

[0050] Parameter determination module, used to establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component;

[0051] A point cloud registration module is used to obtain point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error;

[0052] A feature matching module, configured to determine a feature matching degree based on the point cloud registration error;

[0053] Positioning optimization module, used to obtain positioning optimization targets based on deviation analysis results and engineering constraints;

[0054] An accuracy evaluation module, configured to calculate an assembly accuracy evaluation value based on the positioning optimization target;

[0055] An assembly correction module, configured to determine a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value;

[0056] The comprehensive scoring module is used to calculate a comprehensive quality score based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount, and determine to adopt the corresponding pre-assembly plan when the comprehensive quality score is greater than a preset threshold.

[0057] In addition, to achieve the above-mentioned purpose, the present invention also proposes an electronic device, comprising: a memory for storing a computer software program; a processor for reading and executing the computer software program, thereby realizing a BIM-based digital pre-assembly method for glass curtain walls as described above.

[0058] In addition, to achieve the above-mentioned purpose, the present invention also proposes a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, it implements a BIM-based digital pre-assembly method for glass curtain walls as described above.

[0059] The beneficial effects of the present invention are:

[0060] (1) The present invention realizes accurate identification of on-site installation deviations through a high-precision registration algorithm of point clouds and BIM models; applies an optimization algorithm for spatial coordinate corrections and attitude angle adjustments to effectively improve the positioning accuracy of pre-assembled components; reduces the on-site rework rate caused by mis-installation and incorrect installation, and saves manpower and construction costs.

[0061] (2) The present invention is based on the Internet of Things and a distributed sensor network, and can realize real-time error monitoring and feedback control of the entire assembly process; combined with the assembly accuracy evaluation function and the correction function, it can realize dynamic adaptive adjustment and deviation correction to ensure stable assembly quality.

[0062] (3) The present invention adopts the method of integrating laser scanning and BIM, which effectively solves the problem of complex on-site construction environment and difficulty in integrating multi-source information; the feature matching function combines geometric features, spatial topology and ICP error to achieve deep integration of model and real-scene data; improves the coupling accuracy of construction simulation and real state, and facilitates pre-construction rehearsal and problem prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A flow chart of a BIM-based digital pre-assembly method for glass curtain walls provided by the present invention;

[0064] Figure 2 A structural diagram of a BIM-based digital pre-assembly system for glass curtain walls provided by the present invention;

[0065] Figure 3 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0066] Figure 4 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0068] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0069] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0070] See also Figure 1 , provides a flow chart of a BIM-based glass curtain wall digital pre-assembly method of the present invention, comprising the following steps:

[0071] Step 201: Create a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component.

[0072] During the digital pre-assembly process of the glass curtain wall, a BIM (Building Information Modeling) 3D model of all curtain wall components is first constructed and parametrically designed. This model not only includes the geometric shape of the curtain wall but also embeds the key parameters of the components, including:

[0073] Geometric parameters: such as component length, width, thickness, frame size, etc.;

[0074] Installation location parameters: such as the spatial positioning of the component in the overall curtain wall system (X, Y, Z coordinates), installation angle, orientation, etc.;

[0075] Component identification and logical relationship: such as component number, installation sequence, joint relationship with adjacent components, etc.

[0076] In order to unify the component model and the actual construction coordinate system, the component space coordinate system transformation matrix T can also be defined:

[0077]

[0078] in: is the rotation matrix, which represents the posture of the component relative to the global coordinates; t=[t x ,t y ,t z ] T is the translation vector, which represents the component position offset.

[0079] Step 202: Acquire point cloud data of the construction site through three-dimensional laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error.

[0080] In some embodiments, step 202 may include:

[0081] Each BIM model point is transformed according to the rotation matrix and translation vector to fit each corresponding real-world cloud point to obtain the main registration item;

[0082] Processing the rotation transformation gradient of the rotation matrix and the translation transformation gradient of the translation vector according to the regularization parameter to obtain a regularization term;

[0083] A point cloud registration error is calculated based on the main registration term and the regularization term.

[0084] Wherein, the point cloud registration error is expressed as:

[0085]

[0086] Where E is the point cloud registration error, w i is the weight coefficient, p i and q i is the point pair matched between BIM and point cloud, p i is the BIM model point, q i is the real-world point cloud, λ is the regularization parameter, R is the rotation matrix, t is the translation vector, is the rotation transformation gradient, is the translation transformation gradient.

[0087] In the specific implementation, E is the point cloud registration error, which represents the overall error evaluation function. The smaller the value, the better the registration. i Is the weight coefficient, which indicates the credibility of each point pair and is often used to weaken the influence of noise points.i is a BIM model point, which represents the sampled point in the BIM 3D model. i It is the actual point cloud point, which represents the point cloud of the on-site 3D laser scanning that is related to p i λ is the regularization parameter that controls the influence of the regularization term on the total error (balancing registration accuracy and smoothness). R is the rotation matrix that rotates the BIM model to align with the real-world point cloud. t is the translation vector that translates the BIM model to the position of the real-world point cloud. It is the rotation transformation gradient, which controls the smoothness of the rotation transformation and suppresses drastic rotation changes. Is the gradient of the translation transformation, which controls the smoothness of the translation transformation and suppresses sudden displacement. i +tq i || 2 Describes the Euclidean distance error between the transformed model point and the actual point.

[0088] Main registration term (error minimization kernel) ∑ i w i ||Rp i +tq i || 2 , the goal is to put BIM point p i After transformation R, t, try to fit the corresponding real-world point q i , is a typical least squares point cloud matching (ICP) core term, the weighted term w i Robustness can be improved by calculating point matching quality, reflection intensity, or confidence.

[0089] Regularization term (to prevent overfitting and oscillation) The goal is to prevent drastic changes in R and t, typically measured through first- or second-order derivatives (gradients). This can be understood as introducing a smoothness constraint on the transformation process, making it particularly suitable for large-scale curtain wall point cloud stitching, where it can avoid local misalignment.

[0090] In summary, the minimization of this function in the present invention is usually completed through optimization algorithms (such as Levenberg–Marquardt, Gauss-Newton, ICP), which are used for: automatic alignment of BIM models with measured data, component installation deviation detection, and subsequent optimization analysis (such as input of objective function L and assembly accuracy A).

[0091] Step 203: Determine the feature matching degree based on the point cloud registration error.

[0092] In some embodiments, step 203 may include:

[0093] Obtain the sum of the distances between the BIM model point cloud and the actual point cloud after iterative closest point registration algorithm.

[0094] Obtaining a sum of feature similarities between each of the structures and the corresponding real-scene component in terms of local geometric features, surface normal, reflection intensity, color, and shape;

[0095] Obtaining a sum of topological consistencies indicating consistency in relative positions and connection relationships between each of the structures and the corresponding real-scene components;

[0096] The feature matching degree is obtained by performing a weighted summation on the sum of the point pair distances, the sum of the feature similarities, and the sum of the topological consistency.

[0097] Among them, the feature matching degree is expressed as:

[0098] M=α∑ICP+β∑F+γ∑D;

[0099] Among them, M is the feature matching degree, ICP is the iterative closest point distance, F is the feature similarity, D is the topological consistency, α, β, and γ are the first weight, second weight, and third weight, respectively.

[0100] In the specific implementation, ∑ICP represents the sum of the distances between the BIM model point cloud and the actual point cloud after iterative closest point registration algorithm (such as ICP). It can be expressed as:

[0101] ∑ICP=∑ i ||Rp i +tq i || 2 ;

[0102] It is used to measure the geometric fit between the model and the real scene. It is the most basic registration indicator with high accuracy, but it is sensitive to noise and prone to falling into local optimality.

[0103] ∑F represents the sum of the similarities between BIM components and real-world components in terms of local geometry, surface normal, reflection intensity, color, and shape. Local descriptor algorithms such as FPFH, SHOT, and Spin Image can be used to calculate Euclidean distance, correlation coefficient, and cosine similarity between BIM point clouds and real-world point clouds in feature space. This helps determine whether registration points correspond physically, compensating for the shortcomings of the ICP geometric error term. It also exhibits robustness to non-rigid deformations and complex surfaces.

[0104] D measures whether the spatial relationships between components in the BIM model are consistent with the relative positions and connections between components in the real scene; ∑D is the sum of topological consistency terms. Quantifiable methods include adjacency matrix similarity; spatial orientation matching (up, down, left, right, front, and back); and topological constraints such as distance ratios and angle relationships. This is used to ensure consistency at the overall structural level, and is particularly applicable to complex curtain wall units or special-shaped components, avoiding matching results where the geometric alignment is correct but the assembly relationships are incorrect.

[0105] The sum of α, β, and γ is 1, which can be set based on experience (such as geometry as the main factor and features and topology as the auxiliary factor); or automatically learned through data-driven methods (such as machine learning).

[0106] For example, when the structure of the glass curtain wall is simple: α=0.6, β=0.3, γ=0.1; when the curtain wall is special-shaped or the assembly relationship is complex: α=0.4, β=0.3, γ=0.3.

[0107] In summary, the present invention can quantify the differences between BIM and real scenes from the three levels of "geometry-feature-structure" through the comprehensive matching function M; improve the stability and semantic correctness of the registration, and avoid the problem of "misregistration but minimum distance" when relying solely on ICP; and provide key reference indicators for subsequent positioning optimization (function L), assembly accuracy evaluation (function A), and comprehensive quality scoring (function Q).

[0108] Step 204: Obtain the positioning optimization target based on the deviation analysis results and engineering constraints.

[0109] In some embodiments, step 204 may include:

[0110] Determining a spatial offset error based on an initial position of the point cloud measurement and a target position corresponding to the initial position;

[0111] Determine the attitude deviation based on the attitude angle measured by the point cloud and the target attitude angle corresponding to the attitude angle;

[0112] Determining an error in the seam size based on a seam size measured by the point cloud and a target seam size corresponding to the seam size;

[0113] The spatial offset error, the posture deviation, and the error of the seam size are weightedly summed to obtain the positioning optimization target.

[0114] Among them, the positioning optimization goal can be expressed as:

[0115] L=min{w1∑||P i -P i ′|| 2 +w2∑||θ i -θi ′|| 2 +w3∑||d i -d i ′|| 2};

[0116] Among them, L is the positioning optimization target, P i is the initial position of the point cloud measurement, P i ′ is the target position, θ i is the attitude angle of point cloud measurement, θ i ′ is the target attitude angle, d i is the seam size measured by the point cloud, d i ′ is the target seam size, w1, w2, w3 are the fourth weight, fifth weight, and sixth weight, respectively.

[0117] In specific implementation, the formula minimizes the weighted sum of various errors to obtain the optimal positioning solution that makes the actual installation of the component as close as possible to the design target.

[0118] ||P i -P i ′|| 2 Represents the spatial offset error (translation error) between the actual component position and the target position. ||θ i -θ i ′|| 2 Indicates attitude deviation (such as angle error). ||d i -d i ′|| 2 Indicates error or inconsistency in seam dimensions.

[0119] In some embodiments, w1 can be increased when high spatial accuracy is required, w2 can be increased when the component assembly angle is sensitive, and w3 can be increased when high aesthetics / sealing requirements are required.

[0120] The goal of this formulation is to minimize the total deviation between the measured component state and the design target in the BIM. The result is a set of adjustment parameters (position, angle, spacing) that can be used to guide automated assembly or manual adjustments. Multi-objective optimization algorithms (such as genetic algorithms, particle swarm optimization, and gradient descent) can be used to find the minimum value.

[0121] This function L is directly based on the aforementioned feature matching results (function M) and point cloud deviation analysis results; its output (position, angle, seam adjustment) is used to control the actuator of the IoT assembly system; and the error term δ in the assembly accuracy evaluation function A is calculated. i , ε i ; Ultimately affects the quality scoring function Q.

[0122] In summary, the function of the present invention realizes a high-precision bridge from the static BIM design model to the on-site dynamic assembly execution. Its advantages lie in the integration of multi-dimensional error sources to fully reflect the assembly challenges; flexible adjustment of optimization weights to adapt to different engineering environments; and providing key data support for the realization of digital and high-precision glass curtain wall pre-assembly.

[0123] Step 205: Calculate and obtain an assembly accuracy evaluation value according to the positioning optimization target.

[0124] In some embodiments, step 205 may include:

[0125] Determining the geometric measuring point deviation of each key measuring point according to the measuring point deviation, the corresponding deviation weight and the total number of key measuring points;

[0126] Determine the joint error exponential function according to the correction coefficient, the joint error and the total number of joint measurement points;

[0127] An assembly accuracy evaluation value is calculated based on the geometric measurement point deviation and the joint error exponential function.

[0128] Among them, the assembly accuracy evaluation value can be expressed as:

[0129]

[0130] Among them, A is the assembly accuracy evaluation value, δ i is the measurement point deviation, ε i is the seam error, μ i is the deviation weight, k is the correction coefficient, n is the total number of key measuring points, and m is the total number of joint measuring points.

[0131] In practice, this formula is used to comprehensively evaluate the accuracy level of the glass curtain wall assembly process, evaluating from two dimensions: component measurement point deviation (geometric accuracy) and component joint error (assembly coordination). By weighting and combining these two types of errors, a numerical evaluation result A of the overall assembly quality is obtained. A is the assembly accuracy evaluation value, a numerical indicator that comprehensively evaluates the assembly effect of the entire curtain wall system. The smaller the value, the higher the accuracy. i It is the measurement point deviation, which indicates the error between the actual measured position of the component and the target position (such as position offset, angle deviation). i is the joint error, which represents the deviation between the actual joint width and the target joint width between adjacent components. iis the deviation weight, which indicates the importance of the measurement point in the overall assembly assessment (for example, corner measurement points may be more critical). k is the correction factor, which controls the influence of the second term on the final assembly accuracy score and is typically determined through experience or data training. n is the total number of critical measurement points, which is used to normalize the first term to make it independent of project scale. m is the total number of seam measurement points, which is used to normalize the second term, seam error.

[0132] It is the geometric measuring point deviation, which represents the weighted square deviation average of all key measuring points. It is used to reflect the cumulative position and posture errors during the assembly process and is suitable for monitoring the overall rigidity accuracy.

[0133] This is a negative exponential function of the joint error. The smaller the joint error, the closer the exponential term is to 1, and the greater the contribution to assembly accuracy. Larger joint errors cause this term to decay more rapidly. This helps suppress the amplifying effect of smaller joint errors on the score, reflecting the engineering "tolerance" strategy.

[0134] As you can understand, a smaller A value indicates smaller measurement point and joint errors, indicating high assembly accuracy. Large geometric measurement point deviations and small joint error exponential function terms indicate significant geometric deviations but good joint control, requiring positioning adjustments. Small geometric measurement point deviations and large joint error exponential function terms indicate generally accurate positioning but poor assembly coordination, potentially indicating stretched or overly dense joints. Increasing A indicates severe assembly deviations, potentially requiring recalibration or rework.

[0135] δ in the first term i Derived from the optimization result of the positioning optimization function L; ε in the second term i With seam design size d i ′ and actual installation result d i The output A of this function will be used as: input parameter in correction amount calculation function C; important component of comprehensive quality scoring function Q.

[0136] In summary, the present invention combines weighted averaging with an exponential penalty mechanism to simultaneously evaluate both rigid deviation and flexible error. It also considers the two most critical curtain wall assembly quality indicators, namely, measuring points and joints. This approach can serve as a core quality monitoring indicator for intelligent assembly systems.

[0137] Step 206: Determine a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value.

[0138] In some embodiments, step 206 may include:

[0139] Obtaining a current position deviation representing a currently measured spatial position offset between the component and the target position;

[0140] Calculate the derivative of the accumulated error and determine the error change rate compensation;

[0141] Calculating the integral of the accumulated error to determine an integral correction term;

[0142] The correction amount is determined according to the current position deviation, the error change rate compensation and the integral correction term.

[0143] The correction amount can be expressed as:

[0144]

[0145] Where C is the correction amount, ΔP is the current position deviation, E is the accumulated error, and λ1 and λ2 are the first and second coefficients for controlling the correction sensitivity, respectively.

[0146] In specific implementation, this formula is used to calculate the correction amount C in the pre-assembly process in real time. Its goal is to dynamically feedback and control the installation error, thereby achieving high-precision assembly. It integrates the current position error, the error change rate (trend) and the historical cumulative error to implement an error correction mechanism similar to PID control (proportional-differential-integral). C is the correction amount, which is ultimately used to adjust the correction value of the actuator or operator instruction. The unit is usually length, angle or joint displacement, etc. ΔP is the current position deviation, which is the spatial position offset between the currently measured component and the target position (from the comparison of BIM and point cloud). E is the cumulative error, which is the overall system error at the current moment, and can be the weighted or cumulative value of the deviation over a period of time. is the rate of change of error, indicating the trend of error change (rate of increase / decrease) and can be understood as the "derivative of the error." ∫Edt represents the cumulative effect of error over time and helps systematically adjust for chronic deviations. λ1 and λ2 are the first and second coefficients, respectively, that control the correction sensitivity. The first coefficient controls the effect of the error change rate on the correction amount (response speed), while the second coefficient controls the effect of accumulated error on the correction amount (long-term adjustment capability).

[0147] ΔP is the current position deviation, which represents the spatial offset observed in real time. It is the most direct basis for correction and is often obtained in real time through laser scanning + BIM comparison in actual projects.

[0148] It is the error change rate compensation, similar to the "prediction compensation item", which determines whether the error is getting better or worse. If the error is getting worse (the derivative is positive), the system will actively correct it. It is a key item for controlling response sensitivity.

[0149] λ2∫Edt is an integral correction term, which represents the correction of small errors accumulated over a long period of time. It avoids system offset or systematic drift caused by error accumulation and enhances system stability and anti-interference capability.

[0150] ΔP comes from the position deviation term δ in the assembly accuracy evaluation function A i , E can be defined as the comprehensive deviation time series in the assembly process (such as the output sequence of function A); the output C of this function will participate in: driving the positioning / assembly mechanical system to make real-time adjustments; and at the same time affect the correction term ΔC in the final comprehensive quality scoring function Q.

[0151] Application scenarios can include real-time construction and assembly, intelligent control systems, and error drift monitoring. For example, C can be used to guide robots or operators to dynamically adjust component position and posture, achieving "adaptive calibration" based on error trends. The integral term helps eliminate long-term error accumulation and prevent systematic drift.

[0152] In summary, the present invention constructs an intelligent assembly closed-loop control mechanism with the following advantages: it comprehensively considers current errors, trend judgments, and historical accumulation, making it stable and efficient; it flexibly adjusts the control strategy (through λ1 and λ2) to adapt to different construction precision requirements; and it adapts to the Internet of Things / sensor network system to achieve automated assembly control and precision assurance.

[0153] Step 207: Calculate a comprehensive quality score based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree, and correction amount. When the comprehensive quality score is greater than a preset threshold, determine to adopt a corresponding pre-assembly solution.

[0154] In some embodiments, step 207 may include:

[0155] Determine the assembly accuracy contribution item according to the assembly accuracy evaluation value;

[0156] Determining a point cloud error term according to the point cloud registration error and a maximum value of the point cloud registration error;

[0157] Determining a feature matching item according to the feature matching degree;

[0158] Determine the correction capability item based on the correction amount and the maximum acceptable correction threshold;

[0159] The comprehensive quality score is determined according to the assembly accuracy contribution item, the point cloud error item, the feature matching item, and the correction capability item.

[0160] Among them, the comprehensive quality score can be expressed as:

[0161]

[0162] Among them, Q is the comprehensive quality score, A is the assembly accuracy evaluation value, E is the point cloud registration error, and E max is the maximum value of the point cloud registration error, M is the feature matching degree, C is the correction amount, C maxis the maximum acceptable correction threshold. ω1, ω2, ω3, and ω4 are the seventh, eighth, ninth, and tenth weights, respectively.

[0163] In specific implementation, this formula is a weighted comprehensive evaluation model that normalizes multiple key parameters to unify quality indicators of different dimensions into the same scoring system. Q is the comprehensive quality score, a comprehensive indicator used to evaluate the overall pre-assembly effect of the glass curtain wall, usually standardized to 0-1. A is the assembly accuracy assessment value. E is the point cloud registration error, which comes from the spatial registration error between the laser point cloud and the BIM model, reflecting the degree of deviation between the model and the actual site. E max It is the maximum value of the point cloud registration error, which can be set as the upper limit of the engineering specification or the statistical maximum value, and is used to normalize E. M is the feature matching degree, C is the correction amount, and ΔC represents the degree of position and posture correction in the actual assembly process. C max is the maximum acceptable correction threshold. Exceeding this threshold indicates that the system deviation is too large and the correction effect is out of control. ω1, ω2, ω3, and ω4 are used to adjust the influence of each indicator on the final score.

[0164] ω1A is the contribution item to assembly accuracy, which directly reflects the actual measured component assembly quality. The higher the value, the better. It can be calculated based on the sensor and point cloud errors.

[0165] is the point cloud error term, which normalizes the point cloud registration error E into a relative error; The smaller the error, the higher the score.

[0166] ω3M is a feature matching term that describes the geometric and topological matching between the model and the real scene; a larger value indicates that the BIM is more consistent with the actual situation.

[0167] This is the correction capability item. If the system does not require large corrections (|ΔC| is small), it indicates high initial accuracy and intelligent assembly. This item reflects the automation system's response to and control capabilities for errors.

[0168] Theoretically, Q ranges from 0 (poor quality) to 1 (excellent quality). Adjustability allows for adaptability to different engineering scenarios (e.g., prioritizing geometric error or assembly accuracy) by adjusting the weights. Flexibility combines three metrics: geometric error, structural fit, and assembly control. Therefore, the Q function is the ultimate comprehensive evaluation metric for the performance of the entire pre-assembly system, integrating data from the entire process and optimization results. It serves as a key basis for quality acceptance, feedback control, and system tuning.

[0169] Application scenarios can include construction quality scoring and reporting, using Q as a reference for automatic acceptance or supervision; closed-loop feedback of the assembly system, if the Q score is low, it automatically enters the secondary positioning or reorganization process; system self-learning optimization, Q value feedback can be used to train the weight adjustment or path optimization of the control algorithm.

[0170] In summary, the present invention realizes the standardized integration of multi-dimensional quality factors; emphasizes the triple standards of actual accuracy, error control and model consistency; and supports intelligent construction, closed-loop quality control and data-driven acceptance.

[0171] See also Figure 2 , Figure 2 This is a structural schematic diagram of a BIM-based glass curtain wall digital pre-assembly system provided by the present invention.

[0172] like Figure 2 As shown, an embodiment of the present invention proposes a BIM-based digital pre-assembly system for glass curtain walls, including:

[0173] Parameter determination module 301 is used to establish a parametric BIM three-dimensional model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component;

[0174] The point cloud registration module 302 is used to obtain point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error;

[0175] A feature matching module 303 is configured to determine a feature matching degree based on the point cloud registration error;

[0176] Positioning optimization module 304, used to obtain positioning optimization targets based on deviation analysis results and engineering constraints;

[0177] The accuracy evaluation module 305 is used to calculate the assembly accuracy evaluation value according to the positioning optimization target;

[0178] An assembly correction module 306 is configured to determine a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value;

[0179] The comprehensive scoring module 307 is used to calculate a comprehensive quality score based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount, and determine to adopt the corresponding pre-assembly solution when the comprehensive quality score is greater than a preset threshold.

[0180] It should be noted that, for the specific embodiments of the above modules 301-307 and the beneficial effects brought about, please refer to the above description of steps 201-207, which will not be repeated here.

[0181] See also Figure 3 , Figure 3 Schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented:

[0182] Establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component;

[0183] Acquire point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error;

[0184] Determining a feature matching degree based on the point cloud registration error;

[0185] According to the deviation analysis results and engineering constraints, the positioning optimization target is obtained;

[0186] According to the positioning optimization goal, an assembly accuracy evaluation value is calculated;

[0187] Determining a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value;

[0188] A comprehensive quality score is calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount. When the comprehensive quality score is greater than a preset threshold, the corresponding pre-assembly solution is determined to be adopted.

[0189] See also Figure 4 , Figure 4 Schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented:

[0190] Establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component;

[0191] Acquire point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error;

[0192] Determining a feature matching degree based on the point cloud registration error;

[0193] According to the deviation analysis results and engineering constraints, the positioning optimization target is obtained;

[0194] According to the positioning optimization goal, an assembly accuracy evaluation value is calculated;

[0195] Determining a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value;

[0196] A comprehensive quality score is calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount. When the comprehensive quality score is greater than a preset threshold, the corresponding pre-assembly solution is determined to be adopted.

[0197] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0198] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0199] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0200] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system that is implemented in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0201] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0202] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0203] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A digital pre-assembly method for glass curtain wall based on BIM, characterized in that: The method comprises: Establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component; Acquire point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error; Determining a feature matching degree based on the point cloud registration error; According to the deviation analysis results and engineering constraints, the positioning optimization target is obtained; According to the positioning optimization goal, an assembly accuracy evaluation value is calculated; Determining a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value; A comprehensive quality score is calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount. When the comprehensive quality score is greater than a preset threshold, the corresponding pre-assembly solution is determined to be adopted.

2. The BIM-based digital pre-assembly method for glass curtain wall according to claim 1, characterized in that: The performing multi-source data registration on the point cloud data to calculate the point cloud registration error includes: Each BIM model point is transformed according to the rotation matrix and translation vector to fit each corresponding real-world cloud point to obtain the main registration item; Processing the rotation transformation gradient of the rotation matrix and the translation transformation gradient of the translation vector according to the regularization parameter to obtain a regularization term; A point cloud registration error is calculated based on the main registration term and the regularization term.

3. The BIM-based digital pre-assembly method for glass curtain wall according to claim 2, characterized in that: The point cloud registration error is expressed as: Where E is the point cloud registration error, w i is the weight coefficient, p i and q i is the point pair matched between BIM and point cloud, p i is the BIM model point, q i is the real-world point cloud, λ is the regularization parameter, R is the rotation matrix, t is the translation vector, is the rotation transformation gradient, is the translation transformation gradient.

4. The BIM-based digital pre-assembly method for glass curtain wall according to claim 3 is characterized in that: The determining of the feature matching degree based on the point cloud registration error includes: Obtain the sum of the distances between the BIM model point cloud and the actual point cloud after iterative closest point registration algorithm. Obtaining a sum of feature similarities between each of the structures and the corresponding real-scene component in terms of local geometric features, surface normal, reflection intensity, color, and shape; Obtaining a sum of topological consistencies indicating consistency in relative positions and connection relationships between each of the structures and the corresponding real-scene components; The feature matching degree is obtained by performing a weighted summation on the sum of the point pair distances, the sum of the feature similarities, and the sum of the topological consistency.

5. The BIM-based digital pre-assembly method for glass curtain wall according to claim 4, characterized in that: The feature matching degree is expressed as: M=α∑ICP+β∑F+γ∑D; Among them, M is the feature matching degree, ICP is the iterative closest point distance, F is the feature similarity, D is the topological consistency, α, β, and γ are the first weight, second weight, and third weight, respectively.

6. The BIM-based digital pre-assembly method for glass curtain wall according to claim 5, characterized in that: The positioning optimization target is obtained based on the deviation analysis results and engineering constraints, including: Determining a spatial offset error based on an initial position of the point cloud measurement and a target position corresponding to the initial position; Determine the attitude deviation based on the attitude angle measured by the point cloud and the target attitude angle corresponding to the attitude angle; Determining an error in the seam size based on a seam size measured by the point cloud and a target seam size corresponding to the seam size; The spatial offset error, the posture deviation, and the error of the seam size are weightedly summed to obtain the positioning optimization target.

7. The BIM-based digital pre-assembly method for glass curtain wall according to claim 6, characterized in that: The calculating of the assembly accuracy evaluation value according to the positioning optimization target includes: Determining the geometric measuring point deviation of each key measuring point according to the measuring point deviation, the corresponding deviation weight and the total number of key measuring points; Determine the joint error exponential function according to the correction coefficient, the joint error and the total number of joint measurement points; An assembly accuracy evaluation value is calculated based on the geometric measurement point deviation and the joint error exponential function.

8. The BIM-based digital pre-assembly method for glass curtain wall according to claim 7, characterized in that: The step of determining a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value includes: Obtaining a current position deviation representing a currently measured spatial position offset between the component and the target position; Calculate the derivative of the accumulated error and determine the error change rate compensation; Calculating the integral of the accumulated error to determine an integral correction term; The correction amount is determined according to the current position deviation, the error change rate compensation and the integral correction term.

9. The BIM-based digital pre-assembly method for glass curtain wall according to claim 8, characterized in that: The comprehensive quality score calculated based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount includes: Determine the assembly accuracy contribution item according to the assembly accuracy evaluation value; Determining a point cloud error term according to the point cloud registration error and a maximum value of the point cloud registration error; Determining a feature matching item according to the feature matching degree; Determine the correction capability item based on the correction amount and the maximum acceptable correction threshold; The comprehensive quality score is determined according to the assembly accuracy contribution item, the point cloud error item, the feature matching item, and the correction capability item.

10. A BIM-based digital pre-assembly system for glass curtain walls, characterized in that: The system comprises: Parameter determination module, used to establish a parametric BIM 3D model of the glass curtain wall and determine the geometric parameters and installation position parameters of each component; A point cloud registration module is used to obtain point cloud data of the construction site through 3D laser scanning, perform multi-source data registration on the point cloud data, and calculate the point cloud registration error; A feature matching module, configured to determine a feature matching degree based on the point cloud registration error; Positioning optimization module, used to obtain positioning optimization targets based on deviation analysis results and engineering constraints; An accuracy evaluation module, configured to calculate an assembly accuracy evaluation value based on the positioning optimization target; An assembly correction module, configured to determine a correction amount during the pre-assembly process of the glass curtain wall based on the assembly accuracy evaluation value; The comprehensive scoring module is used to calculate a comprehensive quality score based on the assembly accuracy evaluation value, point cloud registration error, feature matching degree and correction amount, and determine to adopt the corresponding pre-assembly plan when the comprehensive quality score is greater than a preset threshold.

Images