Three-dimensional laser point cloud driven fine modeling method for electric power facility building

Abstract

The invention provides an electric power facility building fine modeling method driven by a three-dimensional laser point cloud. The method comprises the steps that point cloud data are collected, and processing parameters are configured; preprocessing the point cloud data, extracting geometric features to construct an initial model and carrying out geometric accuracy verification, and if not, adjusting registration parameters to carry out modeling again or executing missing region completion; executing structural integrity verification after the verification is passed, terminating modeling if the verification is failed, and performing detail optimization verification if the verification is passed; and if the detail optimization verification is passed, outputting the model, otherwise, configuring an optimization parameter iterative optimization model. The processing parameters comprise the equipment resolution, the scanning angle, the reflection intensity, the geometric topology rule and the like. The verification relates to the standards of curvature distribution, structure size, component connection and the like, and the completion processing adopts a topology reconstruction and feature interpolation method. According to the method, the reliability and the automation level of refined modeling in a complex scene can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital modeling of power facilities, and more particularly, to a three-dimensional laser point cloud driven fine modeling method of power facility buildings. BACKGROUND

[0002] With the continuous development of the power industry, the modeling demand of power facility buildings is increasing. The traditional modeling method mainly relies on manual measurement and two-dimensional drawings, which is low in efficiency and prone to errors. In recent years, three-dimensional laser scanning technology has been gradually applied to the modeling of power facility buildings. By collecting three-dimensional laser point cloud data of buildings, the spatial information can be quickly obtained, providing a more accurate data basis for modeling. However, the existing modeling methods based on three-dimensional laser point cloud still have deficiencies in precision and details.

[0003] The existing three-dimensional laser point cloud modeling method mainly processes point cloud and constructs models simply, lacking strict verification of geometric precision and structural integrity. In the data processing process, only the density and distribution of point cloud are considered, while the reflection intensity and other attributes of point cloud are ignored, resulting in inaccurate expression of key structures and detailed features of the model. In addition, the existing method lacks effective completion and optimization strategies when dealing with point cloud data anomalies. Once there is a lack or anomaly in the point cloud data, the model construction will be severely affected, which cannot meet the requirements of fine modeling of power facility buildings.

[0004] In the implementation of the present application, the inventors have found that the existing methods have at least the following problems or defects: the existing methods do not fully utilize the point cloud attributes, resulting in insufficient geometric precision and detailed feature expression of the model; when the point cloud data is abnormal, there is a lack of effective completion and optimization strategies, which cannot guarantee the structural integrity and accuracy of the model; the existing methods lack strict constraints on engineering specifications during model optimization, which may cause safety hazards in actual application. SUMMARY

[0005] The present application provides a three-dimensional laser point cloud driven fine modeling method of power facility buildings, comprising:

[0006] Collecting three-dimensional laser point cloud data of power facility buildings, configuring data processing parameters according to point cloud attributes;

[0007] The three-dimensional laser point cloud data is preprocessed based on data processing parameters, geometric features are extracted and an initial model is constructed, geometric precision verification is performed to determine whether the initial model meets the preset standard, if the preset standard is not met, the point cloud registration parameters are adjusted, feature extraction and model construction are performed again until the geometric precision verification is passed, if all the preset registration parameter combinations are traversed and the verification is still not passed, it is determined that the point cloud data is abnormal and a missing area completion processing is performed, if the geometric precision verification is passed, a structure integrity verification is performed;

[0008] If the structure integrity verification fails, it is marked as abnormal and the modeling is terminated, if the structure integrity verification is passed, a detail optimization verification is performed, if the detail optimization verification is passed, the final model is output, if the detail optimization verification fails, optimization algorithm parameters are configured and model iterative optimization is performed.

[0009] Further, the data processing parameters include:

[0010] The spatial resolution of the point cloud acquisition device, the scanning angle threshold, the reflection intensity range, and the geometric topology rules required for model construction, the feature matching threshold, and the registration parameter combination, the registration parameter combination includes the point cloud density weight and the curvature constraint coefficient.

[0011] Further, the geometric precision verification includes:

[0012] Compare the model surface curvature distribution with the curvature statistical value of the original point cloud, and determine that the verification fails when the curvature distribution deviation exceeds the preset curvature tolerance;

[0013] Compare the model key structure size with the point cloud measurement value, and determine that the verification fails when the size error absolute value is greater than the preset engineering threshold;

[0014] If both comparisons do not exceed the threshold, it is determined that the geometric precision verification is passed.

[0015] Further, the missing area completion processing includes topology reconstruction and feature interpolation;

[0016] The topology reconstruction includes:

[0017] Generate candidate topology structures for the missing area based on the point cloud data of the adjacent complete area;

[0018] Calculate the geometric continuity index of the candidate topology structure and the complete area;

[0019] Select the topology structure with the optimal geometric continuity index for fusion;

[0020] The feature interpolation includes:

[0021] Extract the spatial distribution rule of the boundary feature points of the missing area;

[0022] constructing a feature propagation function to generate an internal feature point cloud;

[0023] performing a smooth transition between the interpolated point cloud and the original point cloud.

[0024] Further, the structural integrity verification includes:

[0025] configuring integrity detection parameters, including a component connection threshold, a closed surface tolerance, and a topological connectivity standard;

[0026] traversing all structural connection nodes of the model to detect spatial relationships between components;

[0027] calculating gap distances and angular deviations between adjacent surfaces;

[0028] determining that the verification fails when there are unclosed surfaces or broken topological structures;

[0029] determining that the verification passes when all detection items meet the preset standards.

[0030] Further, the detail optimization verification includes feature retention degree verification and engineering specification compliance verification;

[0031] If both the feature retention degree verification and the engineering specification compliance verification pass, the detail optimization verification passes; if either verification fails, the detail optimization verification fails.

[0032] Further, if the feature retention degree verification fails, a feature enhancement algorithm is configured to perform detail restoration;

[0033] If the engineering specification compliance verification fails, a specification constraint condition is configured to perform structural correction; wherein,

[0034] The feature enhancement algorithm includes an edge gradient enhancement amplitude, a feature matching similarity threshold, and a spatial interpolation step length parameter;

[0035] The specification constraint condition includes a minimum cross-sectional width of a component, an upper limit threshold of cantilever length, and a safety clearance standard value.

[0036] Further, the feature enhancement algorithm performing detail restoration includes:

[0037] extracting a high-reflectance intensity feature region in the original point cloud;

[0038] constructing a feature importance evaluation matrix;

[0039] adjusting the geometric details of the corresponding region of the model according to the edge gradient enhancement amplitude;

[0040] verifying the spatial consistency of the enhanced feature and the original point cloud;

[0041] When the consistency deviation exceeds the feature matching similarity threshold, adjust the spatial interpolation step parameter to re-execute the detail recovery.

[0042] Further, the specification constraint condition execution structure modification includes:

[0043] Import the power facility design specification database;

[0044] Establish component size constraint equations and spatial relationship constraint trees;

[0045] Detect safety hazards in the model that violate the minimum cross-sectional width or cantilever length upper threshold of the component;

[0046] Adjust the geometric parameters of the rule-violating component automatically through a constraint satisfaction algorithm;

[0047] Output an optimized model that meets the safety clearance standard value.

[0048] Further, the model iteration optimization includes:

[0049] When the detail optimization verification fails, re-execute the geometric precision verification and the structural integrity verification;

[0050] If the re-verification passes, return to execute the detail optimization verification;

[0051] If the re-verification fails, trigger the missing area completion processing and re-establish the initial model.

[0052] The above embodiments according to the present application at least have the following beneficial effects:

[0053] 1. The precision and reliability of power facility modeling can be improved: by configuring point cloud data processing parameters (such as spatial resolution, reflection intensity range) and combining geometric precision verification (curvature distribution comparison, key structure size checking), point cloud registration errors can be effectively eliminated, ensuring the consistency of model geometric features with original data, and solving the model distortion problem caused by point cloud noise or registration deviation in traditional modeling.

[0054] 2. The integrity of complex structure modeling can be enhanced: based on the missing area completion method of topological reconstruction and feature interpolation, combined with structural integrity verification (component connection detection, surface closure analysis), point cloud missing or broken areas can be automatically repaired, avoiding the limitations of manual intervention, and solving the problem of incomplete model topological structure caused by special-shaped components or occluded areas in power facilities.

[0055] 3. The model can be optimized to adapt to engineering specifications: through detailed optimization verification (feature retention, specification compliance) and dynamic parameter adjustment (feature enhancement algorithm, specification constraint condition), safety hazards in the model (such as cantilever over-limit, insufficient clearance) can be automatically corrected to ensure that the output model meets the power facility design standard, and to solve the problem of low efficiency and standard deviation caused by relying on manual verification in traditional methods. BRIEF DESCRIPTION OF DRAWINGS

[0056] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description read in conjunction with the accompanying drawings. In the drawings, several embodiments of the present application are illustrated by way of example and not limitation in which:

[0057] Figure 1 A flowchart of a three-dimensional laser point cloud driven power facility building fine modeling method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0058] The principles and spirits of the present application will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are given only to enable those skilled in the art to better understand and implement the present application, and do not limit the scope of the present application in any way. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0059] Those skilled in the art know that the embodiments of the present application can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present application can be embodied in the form of a complete hardware, a complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0060] It should be noted that any number of elements in the drawings is used for example and not limitation, and any naming is only used for distinction and does not have any limiting meaning.

[0061] The principles and spirits of the present application will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are given only to enable those skilled in the art to better understand and implement the present application, and do not limit the scope of the present application in any way. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art. Figure 1 Figure 1 A flowchart of a three-dimensional laser point cloud driven power facility building fine modeling method provided by an embodiment of the present application. As shown in Figure 1 The three-dimensional laser point cloud driven power facility building fine modeling method includes:

[0062] S1, collecting three-dimensional laser point cloud data of the power facility building, and configuring data processing parameters according to point cloud attributes;

[0063] ​S2, pre-process the three-dimensional laser point cloud data based on data processing parameters, extract geometric features and construct an initial model, perform geometric precision verification to determine whether the initial model meets the preset standard; if the preset standard is not met, adjust the point cloud registration parameters, re-extract the features and construct the model until the geometric precision verification is passed; if all the preset registration parameter combinations are traversed and still not passed, it is determined that the point cloud data is abnormal and the missing area completion processing is performed; if the geometric precision verification is passed, the structure integrity verification is performed;

[0064] S3, if the structure integrity verification fails, mark the abnormality and terminate the modeling; if the structure integrity verification is passed, perform the detail optimization verification, if the detail optimization verification is passed, output the final model; if the detail optimization verification fails, configure the optimization algorithm parameters and perform the model iterative optimization.

[0065] It should be noted that the present application proposes a three-dimensional laser point cloud driven power facility building fine modeling method, the core of which is to collect three-dimensional laser point cloud data of the power facility building, and configure data processing parameters according to the point cloud attributes, so as to realize fine processing and modeling of the point cloud data. Among them, the three-dimensional laser point cloud data refers to the three-dimensional coordinate point set of the building surface obtained by the laser scanning device, each point contains position information, X, Y, Z coordinates, and reflection intensity and other attributes. Through these data, the three-dimensional model of the building can be constructed. The data processing parameters refer to various parameters that need to be configured when processing the point cloud data, which directly affect the precision of the pre-processing, feature extraction and model construction of the point cloud data. Specifically, these parameters include the spatial resolution of the point cloud acquisition device, the scanning angle threshold, the reflection intensity range, and the geometric topology rules required for model construction, feature matching threshold, registration parameter combination, etc. The reasonable configuration of these parameters is the key to realize high-precision modeling.

[0066] In particular, the spatial resolution in the data processing parameters refers to the minimum distance that the laser scanning device can distinguish, usually measured in millimeters or centimeters, which determines the density and accuracy of the point cloud data. The scanning angle threshold refers to the maximum scanning angle range that the laser scanning device can accept during scanning, used to filter out error data caused by excessive scanning angles. The reflection intensity range refers to the range of reflection intensity values of each point in the point cloud data, which is related to the material and roughness of the object surface. By setting the reflection intensity range, noise points can be filtered out. In addition, the geometric topology rules refer to the rules used to define the geometric shape and topology structure when constructing the model, such as the connection method of edges, the closure of faces, etc. The feature matching threshold refers to the threshold used to judge whether two feature points match during feature extraction, usually a distance or angle threshold. The registration parameter combination refers to the parameter set used in the point cloud data registration process, including point cloud density weight and curvature constraint coefficient, etc., which are used to adjust the alignment accuracy of the point cloud data. The specific settings of these parameters need to be adjusted according to the actual point cloud data and modeling requirements to ensure the accuracy and completeness of the model.

[0067] Preferably, after collecting the three-dimensional laser point cloud data of the power facility building, the step of configuring data processing parameters according to point cloud attributes can be further refined. First, according to the performance of the laser scanning device and the complexity of the building, select an appropriate spatial resolution, for example, for a complex power facility building, the spatial resolution can be set to 1 millimeter to ensure that sufficient detailed point cloud data is obtained. Second, according to the geometric shape of the building and the scanning environment, set the scanning angle threshold, for example, set it to 30 degrees to avoid errors caused by excessive scanning angles. Third, according to the reflection characteristics of the building surface material, set the reflection intensity range, for example, set it to 10 to 255 to filter out noise points. When constructing the initial model, a geometric feature extraction algorithm based on point cloud data can be used, such as calculating the curvature, normal vector, etc. of the point cloud to construct the geometric shape of the model. The input parameters include point cloud data, spatial resolution, scanning angle threshold, reflection intensity range, etc. Through the setting of these parameters, a preliminary three-dimensional model can be generated. When performing geometric accuracy verification, the accuracy of the model can be judged by comparing the curvature distribution of the model surface with the curvature statistical value of the original point cloud, for example, set the curvature tolerance to 0.1, when the curvature distribution deviation exceeds this value, it is determined that the verification fails. Through these refined steps and parameter settings, the accuracy and reliability of the modeling can be improved.

[0068] In some embodiments, the data processing parameters include:

[0069] a spatial resolution of a point cloud acquisition device, a scanning angle threshold, a reflection intensity range, and geometric topology rules required for model construction, feature matching thresholds, and a registration parameter combination including a point cloud density weight and a curvature constraint coefficient.

[0070] It should be noted that the data processing parameters mentioned in the present application are one of the key factors for realizing fine modeling of three-dimensional laser point cloud data. These parameters cover various aspects from data acquisition to model construction, ensuring the accuracy and efficiency of the modeling process. Among them, the spatial resolution of the point cloud acquisition device refers to the minimum distance unit that the laser scanning device can distinguish, which directly affects the density and detail level of the point cloud data; the scanning angle threshold is used to limit the scanning angle range of the scanning device to avoid errors caused by excessive angle; the reflection intensity range is used to filter effective point cloud data and remove noise points. In addition, the geometric topology rules are rules for defining the geometric shape and topological structure of the model, ensuring the rationality of the model structure; the feature matching threshold is used to judge the matching degree between point cloud features; the registration parameter combination includes a point cloud density weight and a curvature constraint coefficient, which is used to optimize the registration process of point cloud data. The reasonable configuration of these parameters is the basis for high-precision modeling.

[0071] Specifically, the spatial resolution of the point cloud acquisition device is usually measured in millimeters (mm) or centimeters (cm), for example, for high-precision power facility modeling, the spatial resolution can be set to 1 millimeter to ensure that sufficient detailed information is obtained. The scanning angle threshold refers to the maximum angle range allowed by the scanning device during scanning, for example, set to 30 degrees, to filter out error data caused by excessive angle. The reflection intensity range refers to the range of reflection intensity values of each point in the point cloud data, usually from 0 to 255, and by setting an appropriate range, background noise and invalid reflection points can be removed. The geometric topology rules include the connection method of edges and the closure of faces, etc., to ensure the rationality of the geometric shape and topological structure of the model. For example, the rules can specify that edges must be connected end-to-end and faces must be closed. The feature matching threshold refers to the threshold used to judge whether two feature points match during feature extraction, which is usually a distance or angle threshold, for example, set to 0.5 millimeters to ensure accurate matching of feature points. The point cloud density weight in the registration parameter combination is used to adjust the density distribution of the point cloud data, for example, set to 1.2 to enhance the weight of high-density areas; the curvature constraint coefficient is used to control the smoothness of curvature changes, for example, set to 0.8 to ensure smooth transition of the model surface. The specific settings of these parameters need to be adjusted according to the actual modeling needs and the characteristics of the point cloud data.

[0072] Preferably, in the configuration of data processing parameters, the setting process of each parameter can be further refined. For example, for the spatial resolution of the point cloud acquisition device, it can be adjusted according to the complexity of the power facility building and the modeling accuracy requirement. If the building structure is complex and high-precision modeling is required, the spatial resolution can be set to 1 millimeter; if the building structure is relatively simple, the spatial resolution can be appropriately relaxed to 5 millimeters. For the scanning angle threshold, it can be adjusted according to the scanning environment and the geometry of the building, for example, set to 20 degrees in a complex environment to reduce the interference of error data. In the setting of the reflection intensity range, it can be adjusted according to the reflection characteristics of the building surface material, for example, for a metal surface, the reflection intensity range can be set to 150 to 255; for a concrete surface, the reflection intensity range can be set to 50 to 150. In the application of geometric topology rules, the structural integrity of the model can be ensured by defining the connection rules of edges and the closure rules of faces. For example, it can be stipulated that the connection angle error of the edge should not exceed 5 degrees, and the closure error of the face should not exceed 0.1 millimeter. In the setting of the feature matching threshold, it can be adjusted according to the distribution density and importance of the feature points, for example, for key structural feature points, the threshold can be set to 0.3 millimeters; for non-key feature points, the threshold can be set to 0.8 millimeters. Through these refined parameter settings, the accuracy and reliability of modeling can be effectively improved, ensuring that the quality of the final model meets the requirements of fine modeling of power facility buildings.

[0073] In some embodiments, the geometric accuracy verification comprises:

[0074] comparing the model surface curvature distribution with the curvature statistical value of the original point cloud, and determining that the verification fails when the curvature distribution deviation exceeds the preset curvature tolerance;

[0075] comparing the model key structure size with the point cloud measurement value, and determining that the verification fails when the size error absolute value is greater than the preset engineering threshold;

[0076] If both comparisons do not exceed the threshold, it is determined that the geometric accuracy verification is passed.

[0077] It should be noted that the geometric accuracy verification is an important link in the present application for ensuring the accuracy of the three-dimensional laser point cloud driven power facility building model. The core is to compare the curvature distribution of the model surface with the curvature statistical value of the original point cloud, and the key structure size of the model with the point cloud measurement value, to judge whether the model meets the preset standard. If the curvature distribution deviation or size error of the model exceeds the preset threshold, it is determined that the verification fails, and the model needs to be adjusted again; if both comparisons do not exceed the threshold, it is determined that the geometric accuracy verification is passed. This process is a key step to ensure that the model can accurately reflect the actual structure of the power facility building. The curvature distribution deviation refers to the difference between the curvature of the model surface and the curvature of the corresponding area in the original point cloud data, and the absolute value of the size error refers to the difference between the size of the key structure in the model and the actual size measured by the point cloud data. By setting a reasonable tolerance range, models that do not meet the accuracy requirements can be effectively screened out.

[0078] Specifically, the curvature distribution deviation refers to the difference between the curvature of the model surface and the curvature of the corresponding area in the original point cloud data. In actual operation, the curvature value of each point on the model surface can be calculated and compared with the curvature value of the corresponding point in the original point cloud data. If the difference between the two exceeds the preset curvature tolerance, it is considered that the curvature distribution of that area does not meet the requirements. For example, the curvature tolerance can be set to 0.05, indicating that the maximum allowed error between the model curvature and the original curvature is 0.05. The absolute value of the size error refers to the difference between the size of the key structure in the model and the actual size measured by the point cloud data. In the verification process, the key structure size of the model needs to be measured and compared with the measurement value in the point cloud data. If the absolute value of the size error is greater than the preset engineering threshold, it is determined that the size does not meet the requirements. For example, the engineering threshold can be set to 1 millimeter, indicating that the maximum allowed error between the model size and the actual size is 1 millimeter. The setting of these parameters needs to be adjusted according to the actual accuracy requirements of the power facility building and the quality of the point cloud data to ensure that the geometric accuracy of the model meets the actual application requirements.

[0079] Preferably, the verification steps and parameter settings can be further refined when performing the geometric accuracy verification. For example, when calculating the curvature distribution of the model surface, a local curvature fitting algorithm can be employed to calculate the curvature values by fitting local regions of the model surface. The input parameters include the point cloud data of the model surface, the size of the fitting region, and the type of fitting algorithm. When comparing the model curvature with the original point cloud curvature, the difference between the two can be calculated point by point, and the number of points whose difference exceeds the curvature tolerance is counted. If the proportion of points exceeding the tolerance exceeds a certain threshold, such as 10%, the curvature distribution verification is determined to fail. When verifying the key structure size, an accurate size measuring tool, such as a laser measuring instrument, can be used to measure the key structure in the model, and the measurement result is compared with the size in the point cloud data. If the absolute value of the size error exceeds the engineering threshold, such as 1 millimeter, the model needs to be adjusted. Through these refined steps and parameter settings, the geometric accuracy of the model can be more accurately verified, ensuring that the model can truly reflect the actual structure of the power facility building.

[0080] In some embodiments, the missing region completion processing includes topology reconstruction and feature interpolation;

[0081] The topology reconstruction includes:

[0082] Generating candidate topologies of the missing region based on point cloud data of adjacent complete regions;

[0083] Calculating geometric continuity indicators of the candidate topologies with the complete regions;

[0084] Selecting the topology with the optimal geometric continuity indicator for fusion;

[0085] The feature interpolation includes:

[0086] Extracting the spatial distribution rule of the boundary feature points of the missing region;

[0087] Constructing a feature propagation function to generate internal feature point cloud;

[0088] Performing smooth transition processing on the interpolated point cloud and the original point cloud.

[0089] It should be noted that the missing area completion processing is a key step in the present application for solving the problem of incomplete three-dimensional laser point cloud data. In actual modeling process, due to occlusion, scanning device performance limitation or environmental interference, etc., missing areas may appear in the point cloud data. In order to ensure the integrity and accuracy of the model, the present application proposes two methods of topology reconstruction and feature interpolation. Topology reconstruction generates candidate topology structures of the missing area and selects the optimal structure for fusion, while feature interpolation extracts the spatial distribution law of the boundary feature points and generates internal feature point cloud to realize smooth transition of the missing area. These methods can effectively fill the missing part of the point cloud data, ensure the geometric continuity and integrity of the model.

[0090] Specifically, topology reconstruction refers to generating candidate topology structures of the missing area based on the point cloud data of the adjacent complete area, and selecting the optimal structure for fusion by calculating the geometric continuity index. Among them, the candidate topology structure refers to the possible topological form of the missing area inferred from the geometric features of the complete area, which can be a plane, a curved surface or a combination of complex shapes. The geometric continuity index is used to measure the geometric matching degree between the candidate topology structure and the complete area, including curvature continuity, normal vector consistency, etc. In feature interpolation, the spatial distribution law of the boundary feature points refers to the geometric features and spatial position relationship of the boundary points of the missing area. By analyzing these laws, a feature propagation function can be constructed. The feature propagation function is used to generate feature point cloud inside the missing area, for example, internal point cloud can be generated according to the distribution law of the boundary points by interpolation algorithm. Smooth transition processing refers to the fusion of the interpolated point cloud and the original point cloud to ensure the geometric continuity between them and avoid obvious splicing marks.

[0091] Preferably, a template matching based method can be employed when performing topology reconstruction. First, a plurality of candidate topology templates are generated according to the geometric characteristics of the adjacent intact region. For example, if the adjacent region is a plane, a plane template is generated; if it is a curved surface, a curved surface template is generated. Then, the geometric continuity index of each template with the intact region is calculated, for example, by calculating the curvature difference and normal vector angle between the template and the intact region to evaluate the matching degree. The template with the optimal geometric continuity index is selected for fusion, for example, the template with the smallest curvature difference and the normal vector angle less than a certain threshold, such as 5 degrees. In the feature interpolation process, a distance weighted interpolation algorithm can be used. First, the spatial distribution rule of the missing region boundary feature points is extracted, for example, the distance and angle relationship between the boundary points is calculated. Then, a feature propagation function is constructed to generate the internal feature point cloud according to the distribution rule of the boundary points. For example, the point cloud density near the boundary can be made higher through distance weighted method, and the point cloud density away from the boundary gradually decreases. Finally, the interpolated point cloud and the original point cloud are processed for smooth transition, for example, by adjusting the density and distribution of the point cloud, the geometric shape and curvature change between the two are smoothly transitioned, so as to realize the complete completion of the missing region.

[0092] In some embodiments, the structural integrity verification comprises:

[0093] configuring integrity detection parameters, the integrity detection parameters comprising member connection threshold, closed surface tolerance and topology connectivity standard;

[0094] traversing all structural connection nodes of the model to detect the spatial relationship between members;

[0095] calculating the gap distance and angular deviation between adjacent surfaces;

[0096] determining that the verification fails when there is an unclosed surface or a broken topology structure;

[0097] determining that the verification passes when all detection items meet the preset standard.

[0098] It should be noted that the structural integrity verification is an important step in the present application for ensuring that the three-dimensional laser point cloud driven power facility building model meets the actual engineering requirements in terms of structural connection and geometric shape. By configuring integrity detection parameters such as component connection threshold, closed surface tolerance and topological connectivity standard, the spatial relationship of all structural connection nodes in the model, the gap distance and angle deviation between adjacent surfaces can be systematically detected. If there is an unclosed surface or broken topology structure, the verification fails; only when all detection items meet the preset standard, the verification is passed. This process is crucial for ensuring the reliability and safety of the model in actual engineering applications. Structural integrity refers to whether the connection relationship and geometric shape between components in the model meet the engineering design requirements, and the integrity detection parameters are specific indicators for quantifying these requirements.

[0099] Specifically, the integrity detection parameters include component connection threshold, closed surface tolerance and topological connectivity standard. Among them, the component connection threshold refers to the maximum value of the allowed connection error between components, for example, for the steel structure connection in the power facility building, the connection error threshold can be set to 1 millimeter, which is used to judge whether the components are correctly connected. The closed surface tolerance refers to the allowed surface closure error range, for example, for the wall surface of the building, the closed tolerance can be set to 0.5 millimeters, which is used to judge whether the surface is closed. The topological connectivity standard refers to whether the topological relationship between each part of the model meets the connectivity requirement, for example, it is required that all components in the model must be connected through continuous edges or surfaces, and there cannot be broken or isolated parts. In the verification process, all structural connection nodes in the model need to be traversed to detect the spatial relationship between components, for example, by calculating the distance and angle between adjacent components to judge whether the connection threshold requirement is met. At the same time, the gap distance and angle deviation between adjacent surfaces are calculated to judge whether they are within the tolerance range. If an unclosed surface or broken topology structure is found, the verification fails, otherwise the verification is passed.

[0100] Preferably, when performing the structural integrity verification, the verification steps and parameter settings can be further refined. For example, when detecting component connections, a geometry feature-based matching algorithm can be used, with input parameters including the geometry of the components, the location of the connection points, and a connection threshold. By calculating the distance and angle between connection points, it is determined whether the components are correctly connected. If the distance exceeds the connection threshold or the angle deviation exceeds a certain set value, such as 5 degrees, the connection node is marked as abnormal. When detecting closed surfaces, a surface closure detection algorithm can be used, with input parameters including the boundary point cloud data of the surface and a closure tolerance. By calculating the distance and angle relationship between the boundary points of the surface, it is determined whether the surface is closed. If there is a distance between the boundary points that exceeds the closure tolerance, the surface is marked as not closed. When detecting topological connectivity, a graph theory algorithm can be used, with the components in the model being regarded as nodes of the graph and the connection relationship being regarded as edges, and by traversing the connectivity of the graph, it is determined whether the model meets the topological connectivity standard. If isolated nodes or broken edges are found, the topological connectivity verification fails. Through these refined steps and parameter settings, the structural integrity of the model can be more comprehensively verified, ensuring the reliability and safety of the model in actual engineering applications.

[0101] In some embodiments, the detail optimization verification includes feature preservation verification and engineering specification compliance verification.

[0102] If both the feature preservation verification and the engineering specification compliance verification pass, the detail optimization verification passes; if either verification fails, the detail optimization verification fails.

[0103] It should be noted that the detail optimization verification is a key link in the present application for ensuring that the three-dimensional laser point cloud driven power facility building model meets high quality standards in terms of feature preservation and engineering specification compliance. This process includes two main parts: feature preservation verification and engineering specification compliance verification. Feature preservation verification is mainly used to evaluate whether the model can accurately reflect the key detail features in the original point cloud data, while engineering specification compliance verification ensures that the geometric shape and structural size of the model comply with the relevant specifications of power facility design. Only when both verification parts pass is the detail optimization verification considered to pass; if either verification fails, the model needs to be further optimized.

[0104] Specifically, the feature retention verification refers to evaluating the retention degree of the model on the details of the original data by comparing the detailed features in the model with the corresponding features in the original point cloud data. For example, the distance deviation between the key feature points in the model and the corresponding points in the original point cloud can be calculated to measure the feature retention. If the deviation exceeds a certain threshold, for example, 0.5 millimeters, it is considered that the feature retention is insufficient. The engineering specification compliance verification refers to checking whether the model meets the requirements of the power facility design specification, such as the minimum cross-sectional width of the component, the upper limit threshold of the cantilever length, and the standard value of the safety clearance. These parameters are set according to the actual use requirements and safety standards of the power facility, to ensure the feasibility and safety of the model in actual engineering applications. For example, the minimum cross-sectional width of the component of the power facility needs to meet certain strength requirements, the cantilever length needs to be within a safe range, and the safety clearance needs to ensure the safe operating distance between equipment.

[0105] Preferably, when performing the detail optimization verification, the verification steps and parameter settings can be further refined. For example, in the feature retention verification, a feature matching algorithm can be used, and the input parameters include the feature point cloud data of the model, the original point cloud data, and the feature matching similarity threshold. By calculating the distance deviation between the feature points of the model and the corresponding points in the original point cloud, the feature retention is evaluated. If the deviation exceeds the feature matching similarity threshold, for example, 0.5 millimeters, the model needs to be processed for detail recovery. In the engineering specification compliance verification, the power facility design specification database can be imported to establish the component size constraint equation and the spatial relationship constraint tree. By detecting whether the size and spatial relationship of the components in the model meet the specification requirements, such as checking whether the minimum cross-sectional width of the component is greater than a certain set value, such as 10 millimeters, whether the cantilever length is less than a certain upper limit threshold, such as 500 millimeters, and whether the safety clearance between equipment meets the standard, such as not less than 200 millimeters. If it is found that the model does not meet the specification requirements, the model needs to be structurally modified. Through these refined steps and parameter settings, the quality of the model in terms of detail retention and engineering specification compliance can be effectively ensured, thereby meeting the needs of the power facility building fine modeling.

[0106] In some embodiments, if the feature retention verification fails, a feature enhancement algorithm is configured to perform detail recovery;

[0107] If the engineering specification compliance verification fails, a specification constraint condition is configured to perform structural modification; wherein,

[0108] The feature enhancement algorithm includes edge gradient enhancement amplitude, feature matching similarity threshold, and spatial interpolation step size parameter;

[0109] The specification constraint condition includes the minimum cross-sectional width of the component, the upper limit threshold of the cantilever length, and the standard value of the safety clearance.

[0110] It should be noted that when the detail optimization verification fails, the application provides targeted optimization strategies to ensure that the model meets the requirements of the power facility building fine modeling in terms of detail features and engineering specification compliance. Specifically, if the feature retention verification fails, a feature enhancement algorithm will be configured to restore the model's detail features; if the engineering specification compliance verification fails, a specification constraint condition will be configured to correct the model structure. This process involves the adjustment and optimization of multiple parameters, aiming to improve the quality of the model through the synergistic effect of algorithms and constraint conditions, so that it meets the requirements of power facility building fine modeling. Among them, the feature enhancement algorithm refers to the expression ability of the detail features in the model through a specific algorithm, and the specification constraint condition refers to the condition for constraining and correcting the model according to the power facility design specification.

[0111] Specifically, the feature enhancement algorithm includes edge gradient enhancement amplitude, feature matching similarity threshold and spatial interpolation step size parameters. Among them, the edge gradient enhancement amplitude refers to the degree of enhancing geometric details in the edge area of the model, and by adjusting this parameter, the key features of the model can be highlighted; the feature matching similarity threshold is used to judge the spatial consistency of the enhanced features and the original point cloud, and when the consistency deviation exceeds the threshold, the parameters need to be adjusted again; the spatial interpolation step size parameter determines the step size of each iteration in the spatial interpolation process, which affects the speed and accuracy of detail recovery. The specification constraint condition includes the minimum cross-sectional width of the component, the upper limit threshold of the cantilever length and the standard value of the safety clearance. These parameters directly correspond to the specific requirements in the power facility design specification, for example, the minimum cross-sectional width of the component ensures that the component has sufficient strength; the upper limit threshold of the cantilever length ensures the stability of the structure; and the standard value of the safety clearance ensures that there is enough safety distance between devices to avoid mutual interference. By reasonably setting these parameters, the model can meet the requirements in terms of both detail features and engineering specification.

[0112] Preferably, in executing the feature enhancement algorithm, the operation steps can be further refined. For example, first, extract high-reflectance intensity feature regions in the original point cloud, which usually correspond to key structural features of the building. Then construct a feature importance evaluation matrix, assigning weights according to the importance of the features. Next, adjust the geometric details of the corresponding regions of the model according to the edge gradient enhancement amplitude, such as increasing the point cloud density in the edge region to enhance the detail performance. After that, verify the spatial consistency of the enhanced features with the original point cloud, if the consistency deviation exceeds the feature matching similarity threshold, for example, 0.3 millimeters, adjust the spatial interpolation step parameter, for example, from 1 millimeter to 0.5 millimeters, and re-execute the detail restoration. In executing the specification constraint condition, the power facility design specification database can be imported to establish component size constraint equations and spatial relationship constraint trees. By detecting components in the model that violate the specifications, such as checking whether the minimum cross-sectional width of a component is less than a set value, such as 10 millimeters, whether the cantilever length exceeds the upper threshold, such as 500 millimeters, and whether the safety clearance between equipment is less than the standard value, such as 200 millimeters. If violations are found, the geometric parameters of the violating components are automatically adjusted by the constraint satisfaction algorithm to ensure that the model meets the safety clearance standard value. Through these refined steps and parameter settings, the detail feature performance and engineering specification compliance of the model can be effectively improved, thus meeting the high-quality requirements of power facility building fine modeling.

[0113] In some embodiments, the feature enhancement algorithm performs detail restoration including:

[0114] extracting high-reflectance intensity feature regions in the original point cloud;

[0115] constructing a feature importance evaluation matrix;

[0116] adjusting the geometric details of the corresponding regions of the model according to the edge gradient enhancement amplitude;

[0117] verifying the spatial consistency of the enhanced features with the original point cloud;

[0118] when the consistency deviation exceeds the feature matching similarity threshold, adjusting the spatial interpolation step parameter to re-execute the detail restoration.

[0119] It should be noted that the execution of the feature enhancement algorithm is a key step in the present application for restoring the model's detailed features. When the feature retention verification fails, the high-reflectance intensity feature regions in the original point cloud are extracted, a feature importance evaluation matrix is constructed, and the geometric details of the corresponding regions of the model are adjusted according to the edge gradient enhancement amplitude, thereby realizing the restoration of details. Verifying the spatial consistency of the enhanced features with the original point cloud is an important step to ensure the effectiveness of the model detail restoration. If the consistency deviation exceeds the feature matching similarity threshold, the spatial interpolation step size parameter needs to be adjusted and the detail restoration needs to be re-executed. This process ensures that the model's detailed features accurately reflect the characteristics of the original point cloud data through dynamic adjustment of the algorithm.

[0120] Specifically, the high-reflectance intensity feature region refers to the region with high reflectance intensity in the original point cloud data. These regions usually correspond to key structural features of buildings, such as edges, corner points, or special material surfaces. By extracting these regions, the positions of details that need to be enhanced can be determined. The feature importance evaluation matrix is a data structure used to evaluate the importance of each feature region. It can assign weights based on the geometric characteristics, reflectance intensity, or other attributes of the features to determine which features need to be enhanced first. The edge gradient enhancement amplitude is a parameter that controls the degree of enhancement of the geometric details of the edge regions of the model. For example, it can be adjusted to increase the sharpness of the edges or highlight specific structures. The spatial interpolation step size parameter determines the step size of each iteration in the spatial interpolation process, affecting the speed and accuracy of detail restoration. If the consistency deviation exceeds the feature matching similarity threshold, for example, set to 0.3 millimeters, the spatial interpolation step size parameter needs to be adjusted, for example, from 1 millimeter to 0.5 millimeter, to restore the details more finely.

[0121] Preferably, when executing the feature enhancement algorithm, the operation steps can be further refined. First, in the original point cloud data obtained by the laser scanning device, the point cloud region with a reflection intensity higher than a certain threshold, for example, 150, is screened out, and these regions are considered as high reflection intensity feature regions. Next, a feature importance evaluation matrix is constructed, and weights are assigned according to the geometric complexity and reflection intensity of the feature regions, for example, a higher weight is assigned to regions with high geometric complexity and strong reflection intensity. Then, the geometric details of the corresponding region are adjusted according to the edge gradient enhancement amplitude adjustment model, for example, the detail performance is enhanced by increasing the point cloud density of the edge region or adjusting the curvature. After that, the spatial consistency of the enhanced feature and the original point cloud is verified, and the consistency is evaluated by calculating the distance deviation between the enhanced feature points and the corresponding points in the original point cloud. If the deviation exceeds the feature matching similarity threshold, the spatial interpolation step parameter is adjusted and the detail restoration is re-executed. For example, the initial step is set to 1 millimeter, and if the consistency deviation is too large, the step is adjusted to 0.5 millimeters to adjust the model details more finely. Through these refined steps and parameter settings, the detail feature performance of the model can be effectively improved, ensuring its consistency with the original point cloud data, thereby meeting the high-quality requirements of power facility building fine modeling.

[0122] In some embodiments, the specification constraint condition performs structure modification, including:

[0123] Importing a power facility design specification database;

[0124] Establishing component size constraint equations and spatial relationship constraint trees;

[0125] Detecting safety hazards in the model that violate the minimum cross-sectional width or cantilever length upper threshold of the components;

[0126] Adjusting the geometric parameters of the violating components automatically through constraint satisfaction algorithms;

[0127] Outputting an optimized model that meets the safety clearance standard value.

[0128] It should be noted that the execution of the specification constraint condition is a key link in the present application for ensuring that the power facility building model meets the engineering design specification. When the engineering specification compliance verification fails, by importing the power facility design specification database, establishing component size constraint equations and spatial relationship constraint trees, detecting components that violate the specification in the model, and automatically adjusting the geometric parameters of the violating components through constraint satisfaction algorithms, an optimized model that meets the safety clearance standard value is finally output. This process ensures the feasibility and safety of the model in actual engineering application through strict specification constraints and automatic parameter adjustment.

[0129] Specifically, the power facility design specification database is a database containing power facility design standards and requirements, which provides specific values and ranges for parameters such as component size, safety clearance, cantilever length, etc. These specifications are developed based on the actual use requirements and safety standards of power facilities to ensure the engineering feasibility of the model. The component size constraint equation is a mathematical relationship that defines the component size must satisfy, for example, the minimum cross-sectional width of the component must be greater than a certain set value, such as 10 mm. The spatial relationship constraint tree is a data structure used to represent the spatial relationship between components, for example, the safety clearance between devices must be greater than a certain standard value, such as 200 mm. By detecting components in the model that violate these constraint conditions, the parts that need to be corrected can be determined. The constraint satisfaction algorithm is an algorithm that automatically adjusts the geometry parameters of the violating components, calculates the new parameter values that meet the requirements according to the specification constraint conditions, and updates the model. The setting of these parameters and algorithms needs to be adjusted according to the specific power facility design specification to ensure the accuracy of the model.

[0130] Preferably, when performing the specification constraint conditions, the operation steps can be further refined. First, import the database containing the power facility design specification, which records the minimum cross-sectional width of the component, the upper threshold of the cantilever length, and the standard value of the safety clearance, etc. Then, according to these specification parameters, establish the component size constraint equation and the spatial relationship constraint tree. For example, for the minimum cross-sectional width of the component, the constraint equation can be set as cross-sectional width ≥ 10 mm; for the safety clearance between devices, the constraint condition can be set as clearance ≥ 200 mm. Then, by traversing all components in the model, it is detected whether there are violations of these constraint conditions. For example, if it is found that the cross-sectional width of a component is less than 10 mm, the geometry parameters are automatically adjusted by the constraint satisfaction algorithm to meet the specification requirements. Specifically, the cross-sectional width value that needs to be increased can be calculated, and the corresponding part in the model is updated. Finally, the optimized model after correction is output to ensure that it fully complies with the power facility design specification. Through these refined steps and parameter settings, the engineering specification compliance of the model can be effectively improved, ensuring its safety and reliability in actual engineering applications.

[0131] In some embodiments, the model iterative optimization includes:

[0132] When the detail optimization verification fails, re-perform the geometric precision verification and the structural integrity verification;

[0133] If the re-verification passes, return to perform the detail optimization verification;

[0134] If the re-verification fails, trigger the missing area completion processing and re-build the initial model.

[0135] It should be noted that model iterative optimization is a key link in the present application for ensuring that the power facility building model can reach high quality standards again after the detail optimization verification fails. When the detail optimization verification fails, the model will re-execute the geometric precision verification and the structural integrity verification to check whether there is a basic problem. If the re-verification passes, return to execute the detail optimization verification to try to solve the problem by adjusting the parameters or optimization algorithm; if the re-verification fails, trigger the missing area completion processing and re-construct the initial model. This process ensures that the model can meet the requirements of fine modeling after multiple optimizations through cyclic iteration.

[0136] Specifically, model iterative optimization refers to the process of gradually improving the quality of the model through multiple verifications and adjustments during the model construction process. Among them, the geometric precision verification is used to check whether the geometric features of the model are consistent with the original point cloud data, including the comparison of curvature distribution and key structural size; the structural integrity verification is used to ensure that the topological structure and component connection relationship of the model meet the engineering requirements, such as checking whether there are unclosed surfaces or broken topological structures. The detail optimization verification focuses on the retention of the detailed features of the model and the compliance with engineering specifications. If problems are found in the detail optimization verification, the model needs to be re-verified for geometric precision and structural integrity to determine the root cause of the problem. If these basic verifications pass, but the detail optimization verification still fails, further optimization adjustment of the model is needed; if the basic verification fails, the missing area needs to be completed and the initial model needs to be re-constructed. This process involves dynamic adjustment of multiple parameters, such as spatial resolution, scanning angle threshold, reflection intensity range, etc., to ensure the final quality of the model.

[0137] Preferably, when performing model iterative optimization, the operation steps can be further refined. First, in the geometric accuracy verification stage, the geometric accuracy of the model can be judged by calculating the deviation between the curvature distribution of the model surface and the curvature statistical value of the original point cloud. If the deviation exceeds the preset curvature tolerance, for example, 0.05, the point cloud registration parameters, such as point cloud density weight or curvature constraint coefficient, need to be adjusted to re-perform feature extraction and model construction. In the structural integrity verification stage, all structural connection nodes of the model can be traversed to detect the spatial relationship between components, such as calculating the gap distance and angle deviation between adjacent surfaces. If unclosed surfaces or broken topological structures are found, topological reconstruction or feature interpolation processing needs to be performed on the model to fill in the missing areas. In the detail optimization verification stage, if the feature retention degree verification fails, the parameters of the feature enhancement algorithm, such as edge gradient enhancement amplitude or spatial interpolation step, can be adjusted; if the engineering specification compliance verification fails, the geometric parameters of the model, such as the minimum cross-sectional width or cantilever length of the component, need to be adjusted according to the specification constraint conditions. Through these refined steps and parameter adjustments, the model can be gradually optimized after multiple iterations to finally meet the high-quality requirements of fine modeling of power facility buildings.

[0138] The above-mentioned various embodiments of the present application have the following beneficial effects:

[0139] 1. The accuracy and reliability of power facility modeling can be improved: by configuring point cloud data processing parameters (such as spatial resolution, reflection intensity range) and combining geometric accuracy verification (curvature distribution comparison, key structural size verification), point cloud registration errors can be effectively eliminated to ensure the consistency of model geometric features with original data, solving the model distortion problem caused by point cloud noise or registration deviation in traditional modeling.

[0140] 2. The integrity of complex structure modeling can be enhanced: based on the missing area completion method of topological reconstruction and feature interpolation, combined with structural integrity verification (component connection detection, surface closure analysis), point cloud missing or broken areas can be automatically repaired to avoid the limitations of manual intervention, solving the problem of incomplete model topological structure caused by special-shaped components or occluded areas in power facilities.

[0141] 3. The adaptability of the model to engineering specifications can be optimized: through detail optimization verification (feature retention degree, specification compliance) and dynamic parameter adjustment (feature enhancement algorithm, specification constraint condition), safety hazards (such as cantilever over-limit, insufficient clearance) in the model can be automatically corrected to ensure that the output model meets the design standards of power facilities, solving the low efficiency and standard execution deviation problem caused by the dependence on manual verification in traditional methods.

[0142] Further, the storage medium of the embodiments of the present application stores program instructions capable of realizing all the methods described above, wherein the program instructions can be stored in the storage medium in the form of a software product, and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media capable of storing program codes, or a computer, a server, a mobile phone, a tablet, and other terminal devices.

[0143] The above description is merely some preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the embodiments of the present application is not limited to the technical solutions formed by the specific combinations of the technical features described above, and should also cover other technical solutions formed by any combinations of the technical features described above or their equivalent features without departing from the inventive concept described above. For example, the technical solutions formed by mutually replacing the above-described features and the technical features disclosed in the embodiments of the present application (but not limited to) having similar functions.

Claims

1. A three-dimensional laser point cloud driven refined modeling method for power facility buildings, characterized in that: The following steps are involved: Collect 3D laser point cloud data of power facility buildings and configure data processing parameters based on point cloud attributes; Preprocess the 3D laser point cloud data based on data processing parameters, extract geometric features and construct an initial model. Perform geometric accuracy verification to determine whether the initial model meets the preset standards. If it does not meet the preset standards, adjust the point cloud registration parameters, re-extract features and re-construct the model until it passes the geometric accuracy verification. If it still fails to pass the verification after traversing all preset registration parameter combinations, the point cloud data is determined to be abnormal and missing area completion is performed. If it passes the geometric accuracy verification, perform structural integrity verification. If the structural integrity verification fails, the exception is marked and the modeling is terminated; if the structural integrity verification passes, the detail optimization verification is performed, and if the detail optimization verification passes, the final model is output; if the detail optimization verification fails, the optimization algorithm parameters are configured and the model iterative optimization is performed.

2. The method according to claim 1, characterized in that The data processing parameters include: The spatial resolution, scanning angle threshold, and reflection intensity range of the point cloud acquisition device, as well as the geometric topology rules, feature matching threshold, and registration parameter combination required for model construction, the registration parameter combination including the point cloud density weight and curvature constraint coefficient.

3. The method according to claim 2, characterized in that The geometric accuracy verification includes: Compare the curvature distribution of the model surface with the curvature statistics of the original point cloud. If the curvature distribution deviation exceeds the preset curvature tolerance, the verification is considered failed. Compare the key structural dimensions of the model with the point cloud measurement values. If the absolute value of the dimensional error is greater than the preset engineering threshold, the verification is considered a failure. If both comparisons do not exceed the threshold, the geometric accuracy verification is determined to have passed.

4. The method according to claim 2, characterized in that The missing region completion process includes topology reconstruction and feature interpolation; The topology reconstruction includes: Generate candidate topological structures of missing regions based on point cloud data of neighboring complete regions; Calculate the geometric continuity index of the candidate topology and the complete region; Select the topological structure with the best geometric continuity index for fusion; The feature interpolation includes: Extract the spatial distribution pattern of feature points at the boundary of the missing area; Construct feature propagation function to generate internal feature point cloud; Perform smooth transition between the interpolated point cloud and the original point cloud.

5. The method according to claim 1, wherein The structural integrity verification includes: Configuring integrity detection parameters, including component connection thresholds, closed surface tolerances, and topological connectivity standards; Traverse all structural connection nodes of the model and detect the spatial relationship between components; Calculate the gap distance and angle deviation between adjacent surfaces; Verification fails when there are unclosed surfaces or broken topology structures; The verification is considered passed when all test items meet the preset standards.

6. The method according to claim 1, characterized in that The detail optimization verification includes feature retention verification and engineering specification compliance verification; If both the feature retention verification and the engineering specification compliance verification are passed, the detail optimization verification is passed; if any one of the verifications fails, the detail optimization verification fails.

7. The method according to claim 6, characterized in that If the feature preservation verification fails, the feature enhancement algorithm is configured to perform detail recovery; If the engineering specification compliance verification fails, the configuration specification constraint conditions are used to perform structural corrections; The feature enhancement algorithm includes edge gradient enhancement amplitude, feature matching similarity threshold and spatial interpolation step size parameters; The code constraints include the minimum cross-sectional width of the component, the upper limit threshold of the cantilever length and the standard value of the safety clearance.

8. The method according to claim 7, characterized in that The feature enhancement algorithm performs detail recovery including: Extract high reflection intensity feature areas from the original point cloud; Construct a feature importance evaluation matrix; Adjust the geometric details of the corresponding area of ​​the model according to the edge gradient enhancement amplitude; Verify the spatial consistency between the enhanced features and the original point cloud; When the consistency deviation exceeds the feature matching similarity threshold, the spatial interpolation step size parameter is adjusted to re-execute detail recovery.

9. The method according to claim 7, characterized in that The specification constraint condition execution structure modification includes: Import the power facility design specification database; Establish component size constraint equations and spatial relationship constraint trees; Detect safety hazards in the model that violate the minimum cross-sectional width or upper limit threshold of cantilever length; Automatically adjust the geometric parameters of the offending components through constraint satisfaction algorithms; Output the optimized model that meets the standard value of safety clearance.

10. The method according to claim 1, characterized in that The model iterative optimization includes: When detail optimization verification fails, re-execute geometric accuracy verification and structural integrity verification; If the re-verification passes, the process returns to the execution details optimization verification; If the revalidation fails, the missing region completion process is triggered and the initial model is rebuilt.

Images