A method, system, processing device and storage medium for detecting diseases and deformations of an LNG storage tank

Abstract

This invention relates to a method, system, processing equipment, and storage medium for detecting defects and deformations in LNG storage tanks. The method includes: acquiring multi-source three-dimensional point cloud data of the LNG storage tank surface; processing the acquired multi-source three-dimensional point cloud data of the LNG storage tank surface to generate a complete high-precision three-dimensional model of the LNG storage tank; obtaining multi-dimensional fusion features of the LNG storage tank based on the generated complete high-precision three-dimensional model; identifying corrosion areas and crack areas on the LNG storage tank surface based on the obtained multi-dimensional fusion features through preset curvature gradient thresholds and texture anomaly patterns to obtain defect identification results of the LNG storage tank; establishing a structural deformation evolution model of the LNG storage tank to generate deformation analysis results of the LNG storage tank; and comprehensively assessing the structural safety status of the LNG storage tank by integrating the defect identification results and deformation analysis results, and outputting a structural safety assessment report of the LNG storage tank. This invention can be widely applied in the field of machine vision technology.

Description

Technical Field

[0001] This invention relates to the field of machine vision technology, and in particular to a method, system, processing equipment, and storage medium for detecting defects and deformation in LNG (liquefied natural gas) storage tanks. Background Technology

[0002] As a core facility for energy storage, the structural health of LNG storage tanks directly impacts operational safety and environmental protection. During long-term use, LNG storage tanks are prone to surface corrosion, cracks, and other localized defects, as well as overall structural deformation, due to factors such as material aging, foundation settlement, temperature stress, and internal pressure circulation. If these defects are not detected and addressed promptly, they can lead to serious leakage accidents. Therefore, regular and accurate defect and deformation detection of LNG storage tanks is crucial.

[0003] Currently, the inspection of LNG storage tanks mainly relies on manual inspection, traditional non-destructive testing (NDT), and machine vision inspection. However, manual inspection methods suffer from low efficiency, long inspection cycles, strong subjectivity, reliance on personnel experience, difficulty in obtaining quantitative data, and high risks associated with working at heights. Furthermore, they struggle to capture macroscopic deformations, as subtle changes are difficult for the human eye to detect. Traditional NDT methods are typically used for precise measurements of specific suspected locations, but they cannot quickly obtain three-dimensional morphological information of the entire tank surface, making it difficult to comprehensively assess the structural condition. They also require close contact, resulting in low efficiency. Machine vision inspection methods, employing 3D laser scanning or wall-climbing robot measurements, suffer from incomplete feature information and a lack of dynamic evolution analysis.

[0004] Therefore, existing technologies cannot achieve integrated, high-precision, and intelligent detection and safety assessment of LNG storage tanks, from microscopic defects to macroscopic deformations, in complex industrial environments. There is an urgent need for a comprehensive detection method that can integrate multi-source three-dimensional data, combine geometric and textural features, and support time-series comparative analysis to overcome the above-mentioned shortcomings. Summary of the Invention

[0005] To address the aforementioned problems, the purpose of this invention is to provide a method, system, processing equipment, and storage medium for detecting defects and deformations in LNG storage tanks, enabling integrated, high-precision, and intelligent detection and safety assessment of LNG storage tanks from microscopic defects to macroscopic deformations.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, it provides a method for detecting defects and deformation in LNG storage tanks, comprising:

[0007] Collect multi-source 3D point cloud data of the LNG storage tank surface; The multi-source 3D point cloud data collected from the surface of the LNG storage tank is processed to generate a complete high-precision 3D model of the LNG storage tank. Based on the generated complete high-precision three-dimensional model, the multi-dimensional fusion features of the LNG storage tank are obtained. Based on the obtained multi-dimensional fusion features, the corrosion area and crack area on the surface of the LNG storage tank are identified by the preset curvature gradient threshold and texture anomaly pattern, and the defect identification results of the LNG storage tank are obtained. Based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in the previous historical cycle, a structural deformation evolution model of the LNG storage tank is established, and deformation analysis results of the LNG storage tank are generated. Based on the results of defect identification and deformation analysis of the LNG storage tank, a comprehensive assessment of the structural safety status of the LNG storage tank is conducted, and a structural safety assessment report of the LNG storage tank is generated.

[0008] Furthermore, the multi-source three-dimensional point cloud data of the LNG storage tank surface includes microscopic morphology point cloud data, macroscopic point cloud data, temperature field monitoring data, and evaporation rate of the LNG storage tank surface. The collection of multi-source three-dimensional point cloud data of the LNG storage tank surface includes: Several lidar devices are installed around the LNG storage tank and on the top platform to collect lidar data from the LNG storage tank. Denoising and filtering preprocessing is performed on lidar data collected by several lidar devices to form macroscopic point cloud data of the overall structure of the LNG storage tank; For suspicious areas and routinely vulnerable areas discovered by lidar scanning, a multi-frequency stripe light sequence was projected using a structured light scanner to scan and to acquire images of deformation stripes modulated on the surface of the LNG tank wall. Using the absolute phase of the highest frequency stripe in the deformed stripe image as a reference, the number of wrapping phases of the medium and low frequency wrapping phases is calculated step by step, and then the absolute phase value of each pixel is calculated. The absolute phase value of each pixel is converted into three-dimensional coordinates through system calibration parameters to generate microscopic topographic point cloud data of the LNG storage tank surface. Simultaneously collect temperature field monitoring data and evaporation rate data from LNG storage tanks.

[0009] Furthermore, the process of processing the multi-source three-dimensional point cloud data collected from the surface of the LNG storage tank to generate a complete, high-precision three-dimensional model that accurately reflects the geometric morphology of the LNG storage tank surface includes: The iterative nearest point algorithm is used to register the micro-topography point cloud data of the LNG tank surface generated by the structured light scanner with the macro-point cloud data of the overall structure of the LNG tank generated by the lidar device, so as to unify the multi-source three-dimensional point cloud data of the LNG tank surface into the same coordinate system. The multi-source 3D point cloud data after unifying the coordinate system are fused to form a point cloud dataset of the LNG storage tank surface. Topological reconstruction and surface reconstruction are performed on the point cloud dataset to generate a complete, high-precision 3D model of the LNG storage tank table.

[0010] Furthermore, based on the generated complete high-precision 3D model, the multi-dimensional fusion features of the LNG storage tank are obtained, including: Extract 3D geometric features from the generated complete high-precision 3D model; Extract two-dimensional texture features from the generated complete high-precision 3D model; The extracted three-dimensional geometric features and two-dimensional texture features are fused to form a multi-dimensional fused feature that can comprehensively reflect the geometric and texture anomalies on the surface of LNG storage tanks: The color texture image is analyzed in the HSV color space. By setting thresholds in the hue and saturation channels, the color abnormality areas caused by corrosion in the color texture image are identified. The extracted 3D geometric features and 2D texture features are fused at the pixel or vertex level to construct a feature vector for each node on the complete high-precision 3D model, forming a multi-dimensional fused feature that can comprehensively reflect the geometric and texture anomalies on the surface of the LNG storage tank.

[0011] Furthermore, based on the obtained multi-dimensional fusion features, and through preset curvature gradient thresholds and texture anomaly patterns, the corrosion areas and crack areas on the surface of the LNG storage tank are identified, resulting in the LNG storage tank defect identification results, including: A pre-defined curvature gradient threshold is used as a geometric condition. When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is linearly distributed, and the color of the area matches the identified color anomaly area, then the area is determined to be a crack area. When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is distributed in a planar manner, and the color of the area matches the identified color anomaly area, then the area is determined to be a corrosion area.

[0012] Furthermore, based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in previous historical cycles, a structural deformation evolution model of the LNG storage tank is established, generating deformation analysis results of the LNG storage tank, including: By comparing the complete high-precision 3D model of the current cycle with the complete high-precision 3D model obtained in the previous historical cycles, the overall deformation and local crack deformation of the model can be obtained. A 3D point cloud distance calculation algorithm is used to calculate the shortest distance from each vertex on the complete high-precision 3D model obtained in the current period to the surface of the complete high-precision 3D model obtained in the previous historical period, and to generate a deformation cloud map. By integrating the complete high-precision three-dimensional model obtained from previous historical cycles, regression analysis is performed on the displacement data of the dome center point and typical wall panel center point of the LNG storage tank to establish a structural deformation evolution model of the LNG storage tank and generate deformation analysis results of the LNG storage tank. If the structural deformation evolution model of the LNG storage tank predicts that the deformation of the LNG storage tank will exceed the preset safety threshold within the next year, a deformation trend warning will be issued.

[0013] Furthermore, based on the comprehensive LNG storage tank defect identification results and deformation analysis results, a comprehensive assessment of the structural safety status of the LNG storage tank is conducted, and a structural safety assessment report of the LNG storage tank is output, including: Spatial mapping of multi-source data is performed to establish a coupled model. The multi-source data includes the LNG storage tank's defect identification results, deformation analysis results, and synchronously collected temperature field monitoring data of the LNG storage tank. By using the established coupling model and integrating multi-source data, the comprehensive risk index of damage to the tank wall insulation layer of LNG storage tanks was calculated. In the complete high-precision 3D model obtained in the current cycle, the disease identification results are highlighted with different colors. The deformation cloud map is overlaid on the complete high-precision 3D model obtained in the current cycle, providing an intuitive visualization output of the deformation distribution; Based on the comprehensive risk index of damage to the tank wall insulation layer of LNG storage tanks, a risk heat map showing the spatial distribution of the risk of damage to the tank wall insulation layer of LNG storage tanks is generated. This heat map is then superimposed on the complete high-precision three-dimensional model obtained in the current cycle as a key output to generate a structural safety assessment report of LNG storage tanks.

[0014] Secondly, a system for detecting defects and deformation in LNG storage tanks is provided, including: The data acquisition module is used to collect multi-source three-dimensional point cloud data of the LNG storage tank surface; The 3D modeling module is used to process the multi-source 3D point cloud data collected from the surface of the LNG storage tank to generate a complete, high-precision 3D model of the LNG storage tank. The feature processing module is used to obtain multi-dimensional fused features of the LNG storage tank based on the generated complete high-precision 3D model. The first analysis and identification module is used to identify corrosion areas and crack areas on the surface of LNG storage tanks based on the obtained multi-dimensional fusion features, through preset curvature gradient thresholds and texture anomaly patterns, and to obtain the defect identification results of LNG storage tanks. The second analysis and identification module is used to establish a structural deformation evolution model of the LNG storage tank based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in the previous historical cycle, and generate deformation analysis results of the LNG storage tank. The diagnostic output module is used to comprehensively assess the structural safety status of LNG storage tanks by integrating the results of defect identification and deformation analysis, and output a structural safety assessment report of the LNG storage tanks.

[0015] Thirdly, a processing device is provided, including a computer program, wherein when the computer program is executed by the processing device, it is used to implement the steps corresponding to the above-mentioned LNG storage tank defect and deformation detection method.

[0016] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, is used to implement the steps corresponding to the above-described LNG storage tank defect and deformation detection method.

[0017] The present invention has the following advantages due to the adoption of the above technical solutions: 1. Achieve integrated comprehensive detection: By integrating multi-source data from structured light (high precision) and lidar (high efficiency), this invention successfully resolves the contradiction between precision and efficiency, enabling simultaneous, efficient, and accurate detection of macroscopic deformation and microscopic defects in storage tanks within the same technical framework.

[0018] 2. Improve the accuracy and robustness of disease identification: This invention deeply integrates three-dimensional geometric features with two-dimensional texture features to construct multi-dimensional fusion features, which overcomes the limitations of single feature information. This enables the system to not only identify cracks with abrupt changes in geometric shape, but also effectively detect corrosion areas with changes in color and texture, greatly reducing the rate of missed detections and false detections.

[0019] 3. Achieving a breakthrough from external observation to internal damage risk assessment: This invention innovatively integrates external three-dimensional topographic data with internal process parameters by constructing a coupled model, making it possible to accurately infer the damage risk of the tank wall insulation layer without entering the tank, greatly expanding the depth and value of detection.

[0020] 4. Achieving a leap from static detection to dynamic safety early warning: This invention compares the current 3D model with historical models over time and establishes a deformation evolution model. This not only assesses the current state of the storage tank but also predicts its deformation development trend, providing key data support for preventive maintenance and realizing true structural health monitoring.

[0021] 5. Automated, intelligent, and visualized testing process: The entire process of this invention is highly automated, which greatly reduces reliance on human labor and errors in subjective judgment. The final output of the three-dimensional annotation model, deformation cloud map, risk heat map, and structured report makes the test results intuitive, quantitative, and easy to understand, greatly improving the efficiency of testing and the scientific nature of decision-making.

[0022] In summary, this invention can be widely applied in the field of machine vision technology. Attached Figure Description

[0023] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. In the drawings: Figure 1 This is a schematic diagram of a method flow provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the system structure provided in an embodiment of the present invention. Detailed Implementation

[0024] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0025] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0026] Although terms such as first, second, third, etc., may be used in this document to describe multiple elements, components, regions, layers, and / or segments, these elements, components, regions, layers, and / or segments should not be limited by these terms. These terms may be used only to distinguish one element, component, region, layer, or segment from another. Unless the context clearly indicates otherwise, terms such as "first," "second," and other numerical terms used herein do not imply order or sequence. Therefore, the first element, component, region, layer, or segment discussed below may be referred to as the second element, component, region, layer, or segment without departing from the teachings of the exemplary embodiments.

[0027] Currently, the inspection of LNG storage tanks mainly relies on manual inspection, traditional non-destructive testing (NDT), and machine vision inspection. However, manual inspection methods suffer from low efficiency, long inspection cycles, strong subjectivity, reliance on personnel experience, difficulty in obtaining quantitative data, and high risks associated with working at heights. Furthermore, they struggle to capture macroscopic deformations, as subtle changes are difficult for the human eye to detect. Traditional NDT methods are typically used for precise measurements of specific suspected locations, but they cannot quickly obtain three-dimensional morphological information of the entire tank surface, making it difficult to comprehensively assess the structural condition. They also require close contact, resulting in low efficiency. Machine vision inspection methods, employing 3D laser scanning or wall-climbing robot measurements, suffer from incomplete feature information and a lack of dynamic evolution analysis. This invention provides a method for detecting defects and deformations in LNG storage tanks, comprising: acquiring multi-source three-dimensional point cloud data of the LNG storage tank surface; processing the acquired multi-source three-dimensional point cloud data of the LNG storage tank surface to generate a complete high-precision three-dimensional model of the LNG storage tank; obtaining multi-dimensional fusion features of the LNG storage tank based on the generated complete high-precision three-dimensional model; identifying corrosion areas and crack areas on the surface of the LNG storage tank based on the obtained multi-dimensional fusion features through preset curvature gradient thresholds and texture anomaly patterns, thereby obtaining defect identification results of the LNG storage tank; establishing a structural deformation evolution model of the LNG storage tank based on the complete high-precision three-dimensional model obtained in the current period and the complete high-precision three-dimensional model obtained in previous historical periods, thereby generating deformation analysis results of the LNG storage tank; and comprehensively evaluating the structural safety status of the LNG storage tank by integrating the defect identification results and deformation analysis results, and outputting a structural safety assessment report of the LNG storage tank. This invention overcomes the problems of low efficiency, small coverage, strong subjectivity, and difficulty in quantification of manual inspection and traditional non-destructive testing methods; it resolves the contradiction between detection accuracy and efficiency in single three-dimensional vision technology, namely, the inability to simultaneously meet the needs of rapid scanning of macroscopic deformation and precise identification of microscopic defects; it overcomes the problem in existing technologies that do not fully utilize feature information, relying only on three-dimensional geometric information while ignoring two-dimensional texture information, resulting in low recognition rates for certain types of defects (such as corrosion with color changes). This invention provides a method, system, processing equipment, and storage medium for detecting defects and deformations in LNG storage tanks that can perform dynamic evolution analysis, enabling early warning of structural deformation trends in LNG storage tanks, rather than just assessment of a single state. Furthermore, this invention can comprehensively assess the potential damage risk to the inner wall insulation layer without entering the tank.

[0028] Example 1 like Figure 1 As shown in the figure, this embodiment provides a method for detecting defects and deformation in LNG storage tanks, including the following steps: 1) Multi-source three-dimensional point cloud data of the LNG storage tank surface are collected by structured light scanning equipment and lidar equipment, including micro-morphological point cloud data, macro-point cloud data, temperature field monitoring data and key process parameters such as evaporation rate of the LNG storage tank surface.

[0029] Specifically, the structured light scanning equipment projects multi-frequency fringe light and acquires deformation fringe images of the LNG tank surface, then uses a multi-frequency phase profile method to analyze and obtain high-precision microscopic topographic point cloud data; the lidar equipment acquires lidar data covering the entire structure of the LNG tank and processes it to obtain macroscopic point cloud data, and simultaneously acquires temperature field monitoring data of the LNG tank wall and key process parameters such as evaporation rate. The specific process of this step is as follows: 1.1) During on-site operations, several lidar devices are installed around the LNG storage tank and on the top platform to ensure that there is an overlap rate of more than 30% between adjacent lidar devices, so as to complete a macroscopic scan of the entire outer wall of the LNG storage tank without blind spots.

[0030] 1.2) Denoising and filtering preprocessing is performed on the lidar data collected by several lidar devices to form macroscopic point cloud data of the overall structure of the LNG storage tank.

[0031] 1.3) For suspicious areas (such as welds, weak areas in previous reports) and routinely vulnerable areas found in lidar scanning, close-range, high-density fine scanning is performed using a structured light scanner. During scanning, the structured light scanner projects a multi-frequency stripe light sequence according to equipment requirements and acquires images of deformed stripes modulated by the LNG tank wall surface.

[0032] 1.4) Using the absolute phase of the highest frequency stripe in the deformed stripe image as a reference, the number of wrapping phases of the medium and low frequency wrapping phases is calculated step by step, and then the absolute phase value of each pixel is calculated.

[0033] 1.5) The absolute phase value of each pixel is converted into three-dimensional coordinates through system calibration parameters to generate microscopic topographic point cloud data of the LNG storage tank surface.

[0034] 2) The multi-source 3D point cloud data collected from the surface of the LNG storage tank is processed to generate a complete, high-precision 3D model that accurately reflects the geometric morphology of the LNG storage tank surface. Specifically: 2.1) The iterative nearest point algorithm is used to register the microscopic topographic point cloud data of the LNG storage tank surface generated by the structured light scanner with the macroscopic point cloud data of the overall structure of the LNG storage tank generated by the lidar device, so as to unify the multi-source three-dimensional point cloud data of the LNG storage tank surface into the same coordinate system.

[0035] Specifically, using macroscopic point cloud data as the global coordinate system reference, the optimal spatial transformation matrix is ​​calculated by automatically finding the nearest point between the two point cloud sets of microscopic morphology point cloud data and macroscopic point cloud data to minimize the error, thereby unifying all microscopic morphology point cloud data under the same global coordinate system.

[0036] 2.2) The multi-source 3D point cloud data after unifying the coordinate system are fused to form a dense point cloud dataset of the LNG storage tank surface: 2.2.1) Voxel mesh filtering is used to downsample the multi-source 3D point cloud data after unifying the coordinate system, which reduces the amount of data and smooths noise while ensuring the accuracy of the model.

[0037] 2.2.2) Determine the overlapping region based on the downsampled multi-source 3D point cloud data.

[0038] 2.2.4) Perform a weighted average on the point cloud data of the overlapping areas to generate a seamless, dense point cloud dataset of the LNG storage tank surface.

[0039] 2.3) Perform topological reconstruction and surface reconstruction on the point cloud dataset of the LNG storage tank surface to generate a complete high-precision 3D model that can accurately reflect the geometric shape of the LNG storage tank surface.

[0040] Specifically, the Poisson surface reconstruction algorithm is used to reconstruct a three-dimensional model from the point cloud dataset of the dense LNG storage tank surface. This algorithm can robustly reconstruct a triangular mesh surface model from the point cloud with directional normals, thus obtaining the complete high-precision three-dimensional model of the present invention. This complete high-precision three-dimensional model not only includes accurate geometric information, but also maps the deformation stripe image collected by the structured light scanning device onto the model surface to form a realistic three-dimensional visualization model.

[0041] 3) Based on the generated complete high-precision 3D model, multi-dimensional fusion features of the LNG storage tank that comprehensively reflect surface geometric anomalies and texture anomalies are obtained, specifically: 3.1) Extract 3D geometric features from the generated complete high-precision 3D model, including but not limited to curvature, curvature gradient distribution, and surface normal vectors: 3.1.1) Based on the generated complete high-precision 3D model, calculate the Gaussian curvature and average curvature of each vertex.

[0042] 3.1.2) The curvature gradient distribution is obtained by calculating the change in curvature values ​​of adjacent triangular facets in the complete high-precision three-dimensional model.

[0043] 3.1.3) Calculate the surface normal vector of each triangular facet in the complete high-precision 3D model. These 3D geometric features are extremely sensitive to identifying surface depressions, protrusions and sharp crack edges.

[0044] 3.2) Extract two-dimensional texture features from the generated complete high-precision three-dimensional model.

[0045] Specifically, a gray-level co-occurrence matrix is ​​used to extract two-dimensional texture features (such as contrast and correlation) from the color texture image mapped onto the surface of the complete high-precision 3D model, which are used to distinguish between normal paint surfaces and texture roughening areas caused by corrosion.

[0046] 3.3) The extracted three-dimensional geometric features and two-dimensional texture features are fused to form a multi-dimensional fused feature that can comprehensively reflect the geometric and texture anomalies on the surface of the LNG storage tank: 3.3.1) Analyze the color texture image in the HSV (H for hue, S for saturation, V for brightness) color space. By setting thresholds in the hue and saturation channels, identify color abnormalities (such as reddish-brown) caused by corrosion in the color texture image, so as to judge the damage through the color abnormality areas.

[0047] 3.3.2) The extracted three-dimensional geometric features and two-dimensional texture features are fused at the pixel or vertex level to construct a feature vector for each node on the complete high-precision three-dimensional model, forming a multi-dimensional fused feature that can comprehensively reflect the geometric anomalies and texture anomalies on the surface of the LNG storage tank. This multi-dimensional fused feature comprehensively characterizes the physical morphology and visual appearance of the sampling point.

[0048] 4) Based on the obtained multi-dimensional fusion features, and through preset curvature gradient thresholds and texture anomaly patterns (such as color anomalies and texture breaks), the corrosion areas and crack areas on the surface of the LNG storage tank are identified, resulting in the LNG storage tank defect identification results, specifically: 4.1) A high curvature gradient threshold (i.e., the curvature gradient of the damaged corrosion area that changes abruptly compared to the normal outer surface of the tank) is pre-set as a geometric condition to find areas with abrupt curvature changes and linear distribution.

[0049] 4.2) When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is linearly distributed, and the color of the area matches the color abnormal area identified in step 3.3.1), then the area is determined to be a crack area.

[0050] 4.3) When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is distributed in a planar manner, and the color of the area matches the color abnormal area identified in step 3.3.1), then the area is determined to be a corrosion area.

[0051] 5) Based on the complete high-precision 3D model obtained in the current cycle and the complete high-precision 3D model obtained in previous historical cycles, a structural deformation evolution model of the LNG storage tank is established, generating deformation analysis results of the LNG storage tank, thereby realizing early warning of the deformation trend of the LNG storage tank, specifically: 5.1) Perform a precise time-series comparison between the complete high-precision 3D model of the current cycle and the complete high-precision 3D model obtained in the previous historical cycles to obtain the overall deformation and local crack deformation of the model.

[0052] 5.2) Using a three-dimensional point cloud distance calculation algorithm, calculate the shortest distance from each vertex on the complete high-precision three-dimensional model obtained in the current period to the surface of the complete high-precision three-dimensional model obtained in the previous historical period, and generate a deformation cloud map using color encoding.

[0053] 5.3) Integrate the complete high-precision three-dimensional model obtained from previous historical cycles, perform regression analysis on the displacement data of the dome center point and typical wall panel center point of the LNG storage tank, establish the structural deformation evolution model of the LNG storage tank, and generate the deformation analysis results of the LNG storage tank (including the deformation cloud map with color coding and the structural deformation evolution model of the LNG storage tank). If the structural deformation evolution model of the LNG storage tank predicts that the deformation of the LNG storage tank will exceed the preset safety threshold within the next year, a deformation trend warning will be issued.

[0054] 6) Based on the comprehensive results of defect identification and deformation analysis of the LNG storage tank, conduct a comprehensive assessment of the structural safety status of the LNG storage tank and output a structural safety assessment report for the LNG storage tank, specifically: 6.1) Spatially map the results of LNG tank defect identification and deformation analysis with the temperature field monitoring data of LNG tank collected simultaneously to establish a coupled model. This model analyzes the correlation between the external deformation area of ​​LNG tank and local low temperature anomalies, and combines the abnormal increase in evaporation rate to comprehensively diagnose the potential damage risk of the tank wall insulation layer of LNG tank.

[0055] Specifically, the process of establishing the coupling model includes: ① Collect multi-source data, including the results of LNG storage tank defect identification, LNG storage tank deformation analysis, and LNG storage tank temperature field monitoring data.

[0056] ② The collected multi-source data is preprocessed and aligned to obtain preprocessed and aligned multi-source data.

[0057] ③ Spatial mapping is performed on the preprocessed and aligned multi-source data to construct a coupled model. This coupled model includes a spatial correlation layer, a logical correlation layer, and a diagnostic decision layer. The spatial correlation layer is used to achieve accurate fusion and correspondence of the preprocessed and aligned multi-source data in the LNG storage tank's three-dimensional spatial coordinate system. The logical correlation layer is used to mine and quantify the inherent logical relationships and statistical patterns among the multi-source data in the spatial correlation layer. The diagnostic decision layer, based on the analysis results of the logical correlation layer and combined with dynamic operating condition data, is used to comprehensively infer and classify the damage risk of the LNG storage tank's wall insulation layer. Spatial mapping involves modeling the LNG storage tank's defect identification results and deformation analysis results into a visualization model, i.e., the coupled model, and mapping the LNG storage tank's temperature field monitoring data to specific locations within the visualization model.

[0058] ④ Validate the constructed coupling model.

[0059] For example, when an abnormal dent or bulge is detected in a specific area of ​​the LNG storage tank wall, and this specific area shows obvious local low temperature anomalies (relative to the surrounding area) on infrared thermography, while the evaporation rate monitoring data shows a continuous abnormal increase, the coupled model will determine that this combination of phenomena is a high-risk feature of damage to the tank wall insulation layer.

[0060] 6.2) Using the established coupling model, multi-source data (including the results of LNG tank defect identification, LNG tank deformation analysis, and LNG tank temperature field monitoring data) are integrated and analyzed to calculate the comprehensive risk index of LNG tank wall insulation layer damage.

[0061] 6.3) In the complete high-precision 3D model obtained in the current cycle, the disease identification results are highlighted with different colors.

[0062] For example, use red to highlight corroded areas and use yellow to highlight cracked areas.

[0063] 6.4) The deformation cloud map is overlaid on the complete high-precision 3D model obtained in the current cycle, providing an intuitive visual output of the deformation distribution.

[0064] 6.5) Based on the comprehensive risk index of the damage to the cold insulation layer of the LNG storage tank wall, generate a risk heat map showing the spatial distribution of the risk of damage to the cold insulation layer of the LNG storage tank wall, and superimpose it as the key output onto the complete high-precision three-dimensional model obtained in the current cycle to generate a structural safety assessment report of the LNG storage tank.

[0065] Specifically, the structural safety assessment report for LNG storage tanks includes a list of defects, detailed location coordinates, maximum deformation and warning levels, as well as a damage risk index of the tank wall insulation layer and maintenance priority recommendations derived from a coupled model.

[0066] Example 2 like Figure 2 As shown in the figure, this embodiment provides an LNG storage tank defect and deformation detection system, including a data acquisition module, a three-dimensional modeling module, a feature processing module, an analysis and identification module, and a diagnostic output module. The analysis and identification module includes a first analysis and identification module and a second analysis and identification module.

[0067] The data acquisition module is used to collect multi-source three-dimensional point cloud data of the LNG storage tank surface using structured light scanning equipment and lidar equipment, respectively.

[0068] The 3D modeling module is used to process multi-source 3D point cloud data collected from the surface of LNG storage tanks to generate a complete, high-precision 3D model that accurately reflects the geometric shape of the LNG storage tank surface.

[0069] The feature processing module is used to obtain multi-dimensional fusion features that can comprehensively reflect surface geometric anomalies and texture anomalies based on the generated complete high-precision 3D model.

[0070] The first analysis and identification module is used to identify corrosion areas and crack areas on the surface of LNG storage tanks based on the obtained multi-dimensional fusion features, through preset curvature gradient thresholds and texture anomaly patterns, and to obtain the defect identification results of LNG storage tanks.

[0071] The second analysis and identification module is used to establish a structural deformation evolution model of the LNG storage tank based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in the previous historical cycle, and generate deformation analysis results of the LNG storage tank.

[0072] The diagnostic output module is used to comprehensively assess the structural safety status of LNG storage tanks based on preset safety procedures, combined with the results of defect identification and deformation analysis, and output a structural safety assessment report for the LNG storage tanks.

[0073] In a preferred embodiment, the LNG storage tank defect and deformation detection system of this embodiment is deployed in a high-performance graphics workstation. It communicates with structured light scanning equipment and lidar equipment through a data interface, controls data acquisition and automatically executes the entire process, and finally delivers a visualized structural safety assessment report of the LNG storage tank.

[0074] In a preferred embodiment, both the structured light scanning device and the lidar device have built-in high-resolution RGB cameras for synchronously acquiring color texture information.

[0075] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0076] Example 3 This embodiment provides a processing device corresponding to the LNG storage tank defect and deformation detection method provided in Embodiment 1. The processing device can be applied to client processing devices, such as mobile phones, laptops, tablets, desktop computers, etc., to execute the method of Embodiment 1.

[0077] The processing device includes a processor, a memory, a communication interface, and a bus. The processor, memory, and communication interface are connected via the bus to enable communication between them. The memory stores a computer program that can run on the processing device. When the processing device runs the computer program, it executes the LNG storage tank defect and deformation detection method provided in Embodiment 1.

[0078] In some implementations, the memory may be high-speed random access memory (RAM), and may also include non-volatile memory, such as at least one disk storage device.

[0079] In other implementations, the processor can be any type of general-purpose processor, such as a central processing unit (CPU) or a digital signal processor (DSP), and there is no limitation here.

[0080] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0081] Those skilled in the art will understand that the structure of the above-described computing device is only a partial structure related to the present invention and does not constitute a limitation on the computing device to which the present invention is applied. A specific computing device may include more or fewer components, or combine certain components, or have different component arrangements.

[0082] Example 4 This embodiment provides a computer program product corresponding to the LNG storage tank defect and deformation detection method provided in Embodiment 1. The computer program product may include a computer-readable storage medium on which computer-readable program instructions for executing the LNG storage tank defect and deformation detection method described in Embodiment 1 are loaded.

[0083] A computer-readable storage medium can be a tangible device that holds and stores instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof.

[0084] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0085] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by a computer program. These computer programs can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that instructions executable by the processor of the computer or other programmable data processing apparatus generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0086] These computer programs may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer programs may also be loaded onto a computer or other programmable data processing equipment, causing a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0088] The above embodiments are only used to illustrate the present invention. The structure, connection method and manufacturing process of each component can be varied. All equivalent transformations and improvements made on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A method for detecting defects and deformation in LNG storage tanks, characterized in that, include: Collect multi-source 3D point cloud data of the LNG storage tank surface; The multi-source 3D point cloud data collected from the surface of the LNG storage tank is processed to generate a complete high-precision 3D model of the LNG storage tank. Based on the generated complete high-precision three-dimensional model, the multi-dimensional fusion features of the LNG storage tank are obtained. Based on the obtained multi-dimensional fusion features, the corrosion area and crack area on the surface of the LNG storage tank are identified by the preset curvature gradient threshold and texture anomaly pattern, and the defect identification results of the LNG storage tank are obtained. Based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in the previous historical cycle, a structural deformation evolution model of the LNG storage tank is established, and deformation analysis results of the LNG storage tank are generated. Based on the results of defect identification and deformation analysis of the LNG storage tank, a comprehensive assessment of the structural safety status of the LNG storage tank is conducted, and a structural safety assessment report of the LNG storage tank is generated.

2. The method for detecting defects and deformation in LNG storage tanks as described in claim 1, characterized in that, The multi-source three-dimensional point cloud data of the LNG storage tank surface includes microscopic morphology point cloud data, macroscopic point cloud data, temperature field monitoring data, and evaporation rate of the LNG storage tank surface. The collection of multi-source three-dimensional point cloud data of the LNG storage tank surface includes: Several lidar devices are installed around the LNG storage tank and on the top platform to collect lidar data from the LNG storage tank. Denoising and filtering preprocessing is performed on lidar data collected by several lidar devices to form macroscopic point cloud data of the overall structure of the LNG storage tank; For suspicious areas and routinely vulnerable areas discovered by lidar scanning, a multi-frequency stripe light sequence was projected using a structured light scanner to scan and to acquire images of deformation stripes modulated on the surface of the LNG tank wall. Using the absolute phase of the highest frequency stripe in the deformed stripe image as a reference, the number of wrapping phases of the medium and low frequency wrapping phases is calculated step by step, and then the absolute phase value of each pixel is calculated. The absolute phase value of each pixel is converted into three-dimensional coordinates through system calibration parameters to generate microscopic topographic point cloud data of the LNG storage tank surface. Simultaneously collect temperature field monitoring data and evaporation rate data from LNG storage tanks.

3. The method for detecting defects and deformation in LNG storage tanks as described in claim 2, characterized in that, The process involves processing the multi-source 3D point cloud data collected from the surface of the LNG storage tank to generate a complete, high-precision 3D model that accurately reflects the geometric morphology of the LNG storage tank surface, including: The iterative nearest point algorithm is used to register the micro-topography point cloud data of the LNG tank surface generated by the structured light scanner with the macro-point cloud data of the overall structure of the LNG tank generated by the lidar device, so as to unify the multi-source three-dimensional point cloud data of the LNG tank surface into the same coordinate system. The multi-source 3D point cloud data after unifying the coordinate system are fused to form a point cloud dataset of the LNG storage tank surface. Topological reconstruction and surface reconstruction are performed on the point cloud dataset to generate a complete, high-precision 3D model of the LNG storage tank table.

4. The method for detecting defects and deformation in LNG storage tanks as described in claim 1, characterized in that, Based on the generated complete high-precision 3D model, the multi-dimensional fused features of the LNG storage tank are obtained, including: Extract 3D geometric features from the generated complete high-precision 3D model; Extract two-dimensional texture features from the generated complete high-precision 3D model; The extracted three-dimensional geometric features and two-dimensional texture features are fused to form a multi-dimensional fused feature that can comprehensively reflect the geometric and texture anomalies on the surface of LNG storage tanks: The color texture image is analyzed in the HSV color space. By setting thresholds in the hue and saturation channels, the color abnormality areas caused by corrosion in the color texture image are identified. The extracted 3D geometric features and 2D texture features are fused at the pixel or vertex level to construct a feature vector for each node on the complete high-precision 3D model, forming a multi-dimensional fused feature that can comprehensively reflect the geometric and texture anomalies on the surface of the LNG storage tank.

5. The method for detecting defects and deformation in LNG storage tanks as described in claim 4, characterized in that, Based on the obtained multi-dimensional fusion features, and through preset curvature gradient thresholds and texture anomaly patterns, corrosion areas and crack areas on the surface of LNG storage tanks are identified, resulting in LNG storage tank defect identification results, including: A pre-defined curvature gradient threshold is used as a geometric condition. When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is linearly distributed, and the color of the area matches the identified color anomaly area, then the area is determined to be a crack area. When the multi-dimensional fusion characteristics of a certain area on the surface of an LNG storage tank exceed the preset curvature gradient threshold, and the curvature of the area is distributed in a planar manner, and the color of the area matches the identified color anomaly area, then the area is determined to be a corrosion area.

6. The method for detecting defects and deformation in LNG storage tanks as described in claim 1, characterized in that, Based on the complete high-precision 3D model obtained in the current cycle and the complete high-precision 3D model obtained in previous historical cycles, a structural deformation evolution model of the LNG storage tank is established, generating deformation analysis results of the LNG storage tank, including: By comparing the complete high-precision 3D model of the current cycle with the complete high-precision 3D model obtained in the previous historical cycles, the overall deformation and local crack deformation of the model can be obtained. A 3D point cloud distance calculation algorithm is used to calculate the shortest distance from each vertex on the complete high-precision 3D model obtained in the current period to the surface of the complete high-precision 3D model obtained in the previous historical period, and to generate a deformation cloud map. By integrating the complete high-precision three-dimensional model obtained from previous historical cycles, regression analysis is performed on the displacement data of the dome center point and typical wall panel center point of the LNG storage tank to establish a structural deformation evolution model of the LNG storage tank and generate deformation analysis results of the LNG storage tank. If the deformation analysis results predict that the deformation of the LNG storage tank exceeds the preset safety threshold within the next year, a deformation trend warning is issued.

7. The method for detecting defects and deformation in LNG storage tanks as described in claim 2, characterized in that, Based on the comprehensive analysis of the defects and deformation of the LNG storage tank, a comprehensive assessment of the structural safety status of the LNG storage tank is conducted, resulting in a structural safety assessment report for the LNG storage tank, including: Spatial mapping of multi-source data is performed to establish a coupled model. The multi-source data includes the LNG storage tank's defect identification results, deformation analysis results, and synchronously collected temperature field monitoring data of the LNG storage tank. By using the established coupling model and integrating multi-source data, the comprehensive risk index of damage to the tank wall insulation layer of LNG storage tanks was calculated. In the complete high-precision 3D model obtained in the current cycle, the disease identification results are highlighted with different colors. The deformation cloud map is overlaid on the complete high-precision 3D model obtained in the current cycle, providing an intuitive visualization output of the deformation distribution; Based on the comprehensive risk index of damage to the tank wall insulation layer of LNG storage tanks, a risk heat map showing the spatial distribution of the risk of damage to the tank wall insulation layer of LNG storage tanks is generated and superimposed on the complete high-precision three-dimensional model obtained in the current period to generate a structural safety assessment report of LNG storage tanks.

8. A system for detecting defects and deformation in LNG storage tanks, characterized in that, include: The data acquisition module is used to collect multi-source three-dimensional point cloud data of the LNG storage tank surface; The 3D modeling module is used to process the multi-source 3D point cloud data collected from the surface of the LNG storage tank to generate a complete, high-precision 3D model of the LNG storage tank. The feature processing module is used to obtain multi-dimensional fused features of the LNG storage tank based on the generated complete high-precision 3D model. The first analysis and identification module is used to identify corrosion areas and crack areas on the surface of LNG storage tanks based on the obtained multi-dimensional fusion features, through preset curvature gradient thresholds and texture anomaly patterns, and to obtain the defect identification results of LNG storage tanks. The second analysis and identification module is used to establish a structural deformation evolution model of the LNG storage tank based on the complete high-precision three-dimensional model obtained in the current cycle and the complete high-precision three-dimensional model obtained in the previous historical cycle, and generate deformation analysis results of the LNG storage tank. The diagnostic output module is used to comprehensively assess the structural safety status of LNG storage tanks by integrating the results of defect identification and deformation analysis, and output a structural safety assessment report of the LNG storage tanks.

9. A processing device, characterized in that, The method includes a computer program, wherein when executed by a processing device, the computer program is used to implement the steps corresponding to the LNG storage tank defect and deformation detection method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it is used to implement the steps corresponding to the LNG storage tank defect and deformation detection method according to any one of claims 1-7.

Images