Installation and construction method for an adiabatic module of a cryogenic storage tank

Abstract

The present invention relates to the technical field of cryogenic storage tank installation, and particularly to an installation construction method for an adiabatic module of a cryogenic storage tank, which includes implanting expansion bolts at predetermined points on the inner wall of the cryogenic storage tank, obtaining the inner wall topography data through three-dimensional laser scanning and generating a point cloud model. Using a BIM system to construct a three-dimensional model of the installation reference plane, calculating the theoretical leveling height at each expansion bolt point, and generating a leveling parameter instruction set. Driving the servo motor of the bolt tightening device to rotate the ball head nut to move the leveling shim to the target height. Generating a pressing path, controlling the pressing robotic arm to align the reserved holes of the adiabatic module with the screw parts of the expansion bolts and press-fitting, so that the surface of the adiabatic module is fully attached to the surface of the leveling shim. Installing fastening nuts on the screw parts of the expansion bolts, performing three-dimensional deviation analysis, and performing iterative correction. Compared with the prior art, the present invention achieves effects such as shortening the construction period, improving the leveling accuracy, reducing stress concentration, and ensuring long-term stability.

Description

Technical Field

[0001] The present invention relates to the technical field of cryogenic storage tank installation, and in particular to an installation construction method for an adiabatic module of a cryogenic storage tank. Background Art

[0002] With the wide application of cryogenic media such as liquefied natural gas (LNG) in the energy field, the safety and reliability of cryogenic storage tanks as key storage facilities have been increasingly emphasized. The cryogenic storage tank insulation system is a core component to ensure the normal operation of the storage tank, and its installation quality directly affects the insulation performance and service life of the storage tank. At present, the installation of cryogenic storage tank adiabatic modules mainly relies on traditional manual methods, which have problems such as low precision, poor efficiency, and unstable quality.

[0003] Traditional methods for installing cryogenic storage tank adiabatic modules usually adopt manual measurement, manual leveling, and fixing methods. First, workers need to drill holes and install expansion bolts at predetermined positions on the inner wall of the concrete outer tank; then, manual measurement and leveling are carried out through tools such as a level; finally, the adiabatic module is installed on the leveled support structure and fixed. This method has large human errors and is difficult to meet the requirements of high-precision installation of the insulation system for modern cryogenic storage tanks.

[0004] With the development of digital technology, three-dimensional laser scanning and BIM (Building Information Modeling) technology have begun to be applied in the field of engineering construction. CN117947954A discloses a high-precision installation method for steel structures based on BIM and three-dimensional laser scanning. This method collects on-site point cloud data through a three-dimensional laser scanner, registers it with the BIM model, and realizes high-precision control of steel structure installation. This method provides a new idea for the precise installation of large structures, but it has not been optimized for the special requirements of cryogenic storage tank adiabatic modules.

[0005] CN106639207A proposes an assembly-based construction method for complex shared space architectural decoration based on BIM. By three-dimensionally scanning the construction site, collecting three-dimensional data and establishing a model, the decoration model is matched to the three-dimensional model for simulated installation and collision testing to find out the non-conforming areas and dimensional deviations. This method has achieved good results in the field of architectural decoration, but it does not involve the special requirements for the installation of cryogenic insulation modules.

[0006] CN115952581A discloses a construction method for the closing node of the ceiling columns in a fine decoration ceiling based on a 3D scan model. This method generates a point cloud model through three-dimensional scanning, performs reverse modeling after integrating with the BIM model, and uses a total station to verify the measured data and feed it back to the 3D information model. This method has reference value for the precise construction of complex structures, but it does not solve the problem of the integration of leveling and fixing in the installation of cryogenic storage tank adiabatic modules.

[0007] CN116911002A introduces an assembly-type computer room virtual pre-assembly method based on BIM and point cloud technology. Point cloud data is obtained through 3D scanning, reverse modeling is carried out and compared with the BIM model for analysis to guide on-site assembly construction. This method is innovative in the field of assembly construction, but it is not designed for the special working conditions of the low-temperature storage tank insulation system.

[0008] CN111709074A proposes a construction method for intelligent control of large-space special-shaped curved surfaces based on BIM technology. By performing 3D laser scanning on the main structure of the civil engineering site, point cloud model data is obtained and processed for analysis, and reverse modeling of the on-site structure is carried out using BIM software. This method has achieved good results in the construction of large-space special-shaped curved surfaces, but it does not consider the precise leveling and fixing requirements for the installation of low-temperature storage tank insulation modules.

[0009] However, the existing technologies still have the following problems in the installation of low-temperature storage tank insulation modules: First, in the traditional installation process, the leveling module and the bolts belong to different processes and need to be positioned and installed multiple times, resulting in a long construction period; second, the leveling module and the insulation module are mostly point contacts (contact area < 5 cm²), and stress concentration is likely to cause structural fatigue; third, the dependence on manual operation is high, and the leveling accuracy is affected by subjective factors, making it difficult to stably meet the industrial standard of ±1 mm; fourth, the existing process lacks digital control means and cannot achieve a full-process closed-loop of "measurement-leveling-fixing".

[0010] Therefore, there is an urgent need for a construction method that can improve the installation accuracy and efficiency of low-temperature storage tank insulation modules while ensuring stable and reliable installation quality. Summary of the Invention

[0011] In order to solve the problems existing in the installation of traditional thin-film storage tank insulation modules, such as process separation, stress concentration, unstable leveling accuracy, lack of digital control, insufficient adaptability, and poor mechanical coupling, and to achieve technical effects such as shortening the construction period, improving the leveling accuracy, reducing stress concentration, and ensuring long-term stability, the present invention provides an installation construction method for low-temperature storage tank insulation modules.

[0012] The object of the present invention can be achieved through the following technical solutions:

[0013] The present invention provides an installation construction method for low-temperature storage tank insulation modules, including the following steps:

[0014] S1. Drill holes at predetermined points on the inner wall of the concrete outer tank of the low-temperature storage tank and implant expansion bolts. Subsequently, obtain the surface topography data of the inner wall of the concrete outer tank of the low-temperature storage tank through 3D laser scanning and generate a point cloud model;

[0015] S2. Based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, calculate the theoretical leveling height at each expansion bolt point, and generate a leveling parameter instruction set;

[0016] S3. According to the leveling parameter instruction set, instruct the servo motor of the bolt tightening device to rotate the ball nut sleeved on the screw part of the expansion bolt, so that the leveling gasket moves along the screw axis to the target height;

[0017] S4. Perform spatial matching between the geometric model of the thermal insulation module and the three-dimensional model of the installation reference plane, generate a pressing path for the thermal insulation module through a path planning algorithm, control the pressing robotic arm to align the reserved hole of the thermal insulation module with the screw part of the expansion bolt, and press it with a preset vertical pressure, so that the surface of the thermal insulation module is fully attached to the surface of the leveling gasket;

[0018] S5. Install a fastening nut on the screw part of the expansion bolt;

[0019] S6. Generate actual flatness point cloud data by three-dimensional laser scanning the installation reference plane, perform three-dimensional deviation analysis with the three-dimensional model of the installation reference plane obtained in S2. If the local error is greater than the preset threshold, use an adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance points, and perform iterative correction through the process of S3.

[0020] Further, in S1, it specifically includes the following steps:

[0021] Drill holes and implant expansion bolts at predetermined points on the inner wall of the concrete outer tank of the cryogenic storage tank to form physical reference points for leveling and fixing;

[0022] Scan the inner wall of the concrete outer tank after implanting the expansion bolts through a three-dimensional laser scanning device to obtain surface topography data, and generate a point cloud model based on the data.

[0023] Further, in S2, it specifically includes the following steps:

[0024] Based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, and optimize the model stiffness distribution by using a finite element analysis algorithm;

[0025] According to the three-dimensional model of the installation reference plane, calculate the theoretical leveling height at each expansion bolt point through a geometric feature extraction algorithm, and generate a leveling parameter instruction set including leveling height, rotation angle, and error threshold.

[0026] Further, in S3, it specifically includes the following steps:

[0027] Based on the leveling parameter instruction set, the target height data of each point is parsed through a motion control algorithm to generate control instructions for the rotation angle, direction, and speed of the servo motor, where the precision of the control instructions matches the lead of the ball screw nut.

[0028] Drive the servo motor of the bolt tightening device to rotate the ball screw nut according to the control instructions, and use a laser rangefinder to monitor the axial displacement of the leveling shim in real time, dynamically correcting the rotation angle until the shim moves to the target height.

[0029] Further, in S4, it specifically includes the following steps:

[0030] Apply resin mastic to the back of the adiabatic module.

[0031] Based on the geometric model of the adiabatic module and the 3D model of the installation reference plane, perform spatial matching through an improved A* path planning algorithm to generate a press-fitting path with obstacle avoidance optimization.

[0032] Control the press-fitting robotic arm to move along the press-fitting path, and use a six-axis force sensor to feedback the pose data in real time, accurately align the reserved hole of the adiabatic module with the screw part of the expansion bolt, and apply a vertical pressure with a holding time of ≥ 30 seconds to make the bottom surface of the adiabatic module and the top surface of the leveling shim form a fully contacting surface.

[0033] Make the resin mastic adhesively connect the pressed adiabatic module to the concrete exterior wall.

[0034] Further, in S4, the specific process of generating a press-fitting path with obstacle avoidance optimization includes:

[0035] Based on the geometric model of the adiabatic module and the 3D model of the installation reference plane, perform spatial coordinate alignment through a point cloud registration algorithm to unify the model coordinate system of the adiabatic module and the model coordinate system of the installation reference plane, and extract the obstacle boundary data of the concave and convex areas on the inner wall of the concrete outer tank.

[0036] In the unified coordinate system, generate a press-fitting path based on the improved A* path planning algorithm. In the improved A* path planning algorithm, the path curvature is optimized through a dynamic weight function to avoid the obstacle boundary and meet the kinematic constraints of the robotic arm, and the path planning accuracy error is ≤ 1 mm.

[0037] Further, in S5, it specifically includes the following steps: Install a fastening nut on the screw part of the expansion bolt with a torque wrench so that each fastening nut is tightened to a preset torque value.

[0038] Further, in S6, it specifically includes the following steps:

[0039] The installation reference plane is scanned in full section by a three-dimensional laser scanner to generate actual flatness point cloud data, and three-dimensional deviation analysis is performed on the actual flatness point cloud data and the three-dimensional model of the installation reference plane in S2. The ICP registration algorithm is used to align the model with the actual point cloud, and the local flatness error value is calculated. If the error ≥ 2 mm / m, it is marked as an out-of-tolerance point position;

[0040] The adaptive leveling algorithm is used to recalculate the leveling compensation amount of the out-of-tolerance point position, and iterative correction is performed through the process of S3.

[0041] Furthermore, the specific process of correction includes:

[0042] Subsequently, based on the coordinates of the out-of-tolerance point position, the adaptive leveling algorithm is used to recalculate the leveling compensation amount. The adaptive leveling algorithm iteratively optimizes the compensation amount ΔH by the gradient descent method, and the calculation formula is:

[0043] ΔH = k·(H 实际 - H 理论 ), where H 实际 is the actual measured height, H 理论 is the theoretical design height,

[0044] k is the convergence coefficient, and the value range is 0.8 - 1.2,

[0045] and the corrected leveling parameter instruction set is fed back to the servo motor control process of S3, and the ball head nut is driven by the servo motor for secondary leveling until the error is less than the preset threshold.

[0046] Furthermore, in S3, the output end of the servo motor of the bolt tightening device can be connected to the ball head nut through a rotating kit to achieve torque transmission.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1) Aiming at the problem that in the traditional installation process, the leveling module and the bolts belong to different processes and multiple positioning and installations are required, resulting in a long construction period, the present invention integrates a composite structure (ball head nut + leveling gasket) with leveling and fixing functions, combines the traditional split construction processes into a single operation, reduces the manual intervention link by more than 40%, and significantly shortens the construction period.

[0049] 2) Aiming at the problem that the leveling module and the heat insulation module are mostly point contacts and stress concentration is likely to cause structural fatigue, the present invention adopts the full-area contact design of the leveling gasket and the heat insulation module, and cooperates with the elastic pre-tightening force mechanism of the disc spring, which can reduce the local stress concentration to less than 20% of the traditional point contact mode, effectively avoiding structural fatigue damage.

[0050] 3) Aiming at the problems of high dependence on manual operation and difficulty in stably meeting the industrial standard of ±1mm for leveling accuracy due to subjective factors, the present invention combines a closed-loop control system driven by three-dimensional laser scanning and BIM model, improving the leveling accuracy to ±0.3mm, with the flatness error of the installation reference plane <2mm / m, which is more than 3 times higher than the prior art (±1mm).

[0051] 4) Aiming at the problems that the existing process lacks digital control means and cannot achieve a full-process closed loop of "measurement - leveling - fixing", the present invention realizes a full-process digital closed-loop control from measurement to leveling and then to fixing through an adaptive leveling algorithm and a path planning algorithm. Through real-time monitoring and feedback correction, the accuracy and stability of the installation process are ensured, and the reliability and consistency of the installation quality are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic flow chart of the installation construction method of the cryogenic storage tank insulation module in the present invention.

[0053] Figure 2 It is a schematic structural diagram of the specific installation of the expansion bolt in the present invention.

[0054] Figure 3 It is a schematic side sectional view of the installation reference plane in the present invention.

[0055] Figure 4 It is a schematic top sectional view of the installation reference plane in the present invention.

[0056] Figure 5 It is a schematic diagram of the setting of the resin mortar in the present invention.

[0057] In the figure: 1. Expansion bolt, 2. Fastening nut, 3. Leveling gasket, 4. Insulation module, 5. Concrete outer wall, 6. Ball nut, 7. Resin mortar. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] The present invention will be described in detail below with reference to the drawings and specific embodiments. Features such as component models, material names, connection structures, control methods, algorithms, etc. that are not clearly described in the present technical solution are regarded as common technical features disclosed in the prior art.

[0059] Embodiment 1

[0060] See Figure 1 , in this embodiment, an installation construction method for a cryogenic storage tank insulation module includes the following steps:

[0061] S1: Drill holes at predetermined points on the inner wall of the concrete outer tank of the cryogenic storage tank and implant the expansion bolt 1. Subsequently, obtain the surface topography data of the inner wall of the concrete outer tank of the cryogenic storage tank through three-dimensional laser scanning to generate a point cloud model;

[0062] Specifically, drill holes at predetermined points on the inner wall of the concrete outer tank of the cryogenic storage tank and implant expansion bolts 1 to form physical reference points for leveling and fixing; use a three-dimensional laser scanning device to scan the inner wall of the concrete outer tank after implanting the expansion bolts 1, obtain surface topography data, and generate a point cloud model based on the data.

[0063] In this embodiment, first, according to the design drawings of the cryogenic storage tank and the installation requirements of the insulation module, determine the layout positions of the expansion bolts 1. Usually, points are arranged on the inner wall of the concrete outer tank according to the principle of uniform distribution, and the point density is 4 - 6 points per square meter. Use an industrial-grade electric drill to drill holes at the predetermined points, with a drill hole diameter of 16 mm and a depth of 120 mm. After drilling, clean the debris in the holes, implant expansion bolts 1 of M12 specification, and use a torque wrench to initially fix the expansion bolts 1, with the torque controlled at 80 N·m.

[0064] After implanting the expansion bolts 1, use a FARO Focus S350 three-dimensional laser scanner to perform a full-range scan of the inner wall of the concrete outer tank. During the scanning process, set the scanner at the center position of the storage tank and rotate 360 degrees for scanning, with the scanning accuracy set at ±2 mm. To ensure the integrity of the scanned data, set multiple scanning stations inside the storage tank, and the overlapping area between adjacent stations is not less than 30%. After scanning, process the original scanned data through FARO SCENE software, including noise filtering, point cloud registration, and data fusion, and finally generate a complete point cloud model of the surface topography of the inner wall of the concrete outer tank, with a point cloud density reaching 1000 points per square centimeter.

[0065] S2: Based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, calculate the theoretical leveling height at each point position of the expansion bolts 1, and generate a leveling parameter instruction set.

[0066] Specifically, based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, where a finite element analysis algorithm is used to optimize the model stiffness distribution; according to the three-dimensional model of the installation reference plane, calculate the theoretical leveling height at each point position of the expansion bolts 1 through a geometric feature extraction algorithm, and generate a leveling parameter instruction set including leveling height, rotation angle, and error threshold.

[0067] In this embodiment, the acquired point cloud model is imported into the Autodesk Revit BIM system, and an accurate three-dimensional model of the inner wall of the concrete outer tank is constructed through the point cloud to BIM conversion module. During the construction process, the system automatically identifies the positions of the expansion bolts 1 and marks them. Based on the model of the inner wall of the concrete outer tank, the ANSYS finite element analysis software is used to optimize the stiffness distribution, establish a mesh model with an element size of 10 mm × 10 mm, and consider the elastic modulus and Poisson's ratio of the concrete material to be 30 GPa and 0.2 respectively. Through finite element analysis, the weak areas and the areas with better rigidity of the inner wall structure are identified, and the support point distribution of the thermal insulation module is optimized to ensure that the installation reference surface has sufficient stiffness and stability.

[0068] Based on the optimized model, an ideal installation reference surface is constructed. This reference surface is a theoretical plane representing the target plane for the installation of the thermal insulation module. Using the geometric feature extraction algorithm, the vertical distance from each expansion bolt 1 position to the ideal reference surface, that is, the theoretical leveling height, is calculated. The algorithm uses the least squares method for fitting to ensure the minimum overall deviation between the reference surface and the actual inner wall. For each bolt position, an accurate leveling height value is calculated with an accuracy of 0.1 mm. At the same time, the rotation angle required to adjust the ball head nut 6 is calculated. Considering the lead of the M12 thread is 1.75 mm / rev, the height value is converted into a rotation angle value. Finally, a leveling parameter instruction set is generated, including the coordinates of each position, the target leveling height, the required rotation angle, and the allowable error threshold (±0.5 mm). These parameters are saved as a structured data file for subsequent automated leveling control.

[0069] S3: According to the leveling parameter instruction set, instruct the servo motor of the bolt tightening device to rotate the ball head nut 6 sleeved on the screw part of the expansion bolt 1, so that the leveling gasket 3 moves axially along the screw to the target height;

[0070] Specifically, based on the leveling parameter instruction set, the target height data of each position is analyzed through the motion control algorithm to generate the rotation angle, direction, and speed control instructions of the servo motor, where the control instruction accuracy matches the lead of the ball head nut 6; drive the servo motor of the bolt tightening device to rotate the ball head nut 6 according to the control instructions, and the axial displacement of the leveling gasket 3 is monitored in real time through a laser rangefinder, and the rotation angle is dynamically corrected until the gasket moves to the target height.

[0071] In this embodiment, the output end of the servo motor of the bolt tightening device can be connected to the ball head nut 6 through a rotating kit to achieve torque transmission. The rotating kit is a commonly used tool for those skilled in the art and will not be elaborated here.

[0072] In this embodiment, first, the leveling parameter instruction set is imported into the control system of the bolt tightening device. The control system uses a Siemens S7-1200 PLC as the core controller, in cooperation with a Mitsubishi MR-J4 series servo drive and servo motor. The control system analyzes the target height data of each point through a motion control algorithm, and converts the height value into precise rotation parameters of the servo motor. Considering that the lead of the M12 thread is 1.75 mm / rev, the system divides the target height by the lead value to calculate the required number of rotation turns. For precise control, the servo motor uses a 2000-line incremental encoder with a resolution of 0.018 degrees / pulse, ensuring that the rotation angle control accuracy is higher than 0.1 degree.

[0073] During specific implementation, the mechanical structure of the bolt tightening device includes a servo motor, a reducer, and a special rotation kit. The rotation kit includes a hexagonal sleeve structure similar to a hoop. The hexagonal sleeve structure is a one-sided hinged and openable structure, and on the other side, there is a fixing piece (the fastening principle is similar to a hoop) that can be fastened by bolts. Its inner surface is a hexagonal structure that can match the ball head nut 6, and it can quickly connect with the ball head nut 6 and transmit torque. The rated torque of the servo motor is 5 N·m, and a 10:1 reducer provides sufficient output torque. During operation, the operator puts the rotation kit on the ball head nut 6 and starts the automatic leveling program. The servo motor operates according to the preset rotation angle, direction, and speed parameters. The initial speed is set at 60 rpm and automatically reduces to 10 rpm when approaching the target position to ensure precise positioning.

[0074] During the leveling process, the system uses a laser rangefinder installed on the rotation kit to monitor the axial displacement of the leveling shim 3 in real time. The measurement accuracy of the laser rangefinder is ±0.05 mm, and the sampling frequency is 100 Hz. The control system compares the displacement data measured in real time with the target height, and dynamically adjusts the rotation angle of the servo motor through a PID control algorithm until the shim moves to the target height. When the deviation between the measured height and the target height is less than ±0.2 mm, the system determines that the leveling is completed, automatically stops the servo motor, and records the actual leveling height data.

[0075] S4: Apply the resin mortar 7 on the back of the thermal insulation module 4. Refer to Figure 5 , in this technical solution, the back of the thermal insulation module 4 is defined as the direction facing the concrete exterior wall. The application of the resin mortar 7 should conform to the preset construction rules. Match the geometric model of the thermal insulation module with the three-dimensional model of the installation reference surface, generate the pressing path of the thermal insulation module 4 through a path planning algorithm, control the pressing robotic arm to align the reserved hole of the thermal insulation module 4 with the screw part of the expansion bolt 1, and press it with a preset vertical pressure. Refer to Figure 4 , so that the surface of the thermal insulation module 4 is fully attached to the surface of the leveling shim 3. Figure 2Only the external plate for installation of the adiabatic module 4 is shown, that is Figure 2 the adiabatic module 4 shown in Figure 2 . This part, as the external plate of the adiabatic module 4, is connected to the concrete exterior wall 5 by expansion bolts 1, so that the resin mastic 7 adhesively connects the pressed adiabatic module to the concrete exterior wall.

[0076] Specifically, based on the geometric model of the adiabatic module 4 and the three-dimensional model of the installation reference plane, the improved A* path planning algorithm is used for spatial matching to generate a press-fitting path with obstacle avoidance optimization. Control the press-fitting robotic arm to move along the press-fitting path, and the pose data is real-time fed back through a six-axis force sensor. Align the reserved holes of the adiabatic module 4 precisely with the screw part of the expansion bolt 1, and apply a vertical pressure. The pressure holding time is ≥ 30 seconds, so that the bottom surface of the adiabatic module 4 and the top surface of the leveling shim 3 form a full-contact surface.

[0077] Specifically, before press-fitting the adiabatic module, resin glue needs to be applied to the back of the adiabatic module 4 according to the design requirements. Its function is to firmly bond the adiabatic module 4 to the concrete exterior wall 5 after it solidifies, play a role in evenly transmitting the load, and avoid all loads such as liquid cargo pressure being solely borne by the expansion bolt 1.

[0078] In this embodiment, the specific process of generating a press-fitting path with obstacle avoidance optimization includes: based on the geometric model of the adiabatic module 4 and the three-dimensional model of the installation reference plane, the spatial coordinate alignment is performed through the point cloud registration algorithm to unify the coordinate systems of the adiabatic module model and the installation reference plane model, and the obstacle boundary data of the concave and convex areas on the inner wall of the concrete outer tank is extracted. See the schematic diagram of the installation reference plane in Figure 3 and Figure 4 ; in the unified coordinate system, a press-fitting path is generated based on the improved A* path planning algorithm. In the improved A* path planning algorithm, the path curvature is optimized through a dynamic weight function to avoid the obstacle boundary and meet the kinematic constraints of the robotic arm. The path planning accuracy error is ≤ 1 mm.

[0079] In this embodiment, first, the CAD geometric model of the adiabatic module 4 is imported into the control software of the press-fitting system. The adiabatic module 4 is usually a rectangular block with a size of about 2400 mm × 1200 mm × 350 mm, which is composed of two layers of wooden plywood and the middle polyurethane foam. The model contains the accurate position information of the reserved holes. The diameter of the reserved holes is 18 mm, which is 2 mm larger than the screw diameter of the expansion bolt 1 for easy installation.

[0080] Use the point cloud registration algorithm to align the spatial coordinates of the adiabatic module 4 model with the three-dimensional model of the installation reference plane. The iterative closest point (ICP) algorithm is used in the registration process, with the maximum number of iterations set to 50 and the convergence threshold set to 0.001 mm. After registration, the two coordinate systems are unified to form a unified working space. At the same time, the system automatically extracts the concave and convex regions on the inner wall of the concrete outer tank as obstacle boundary data to establish an obstacle map.

[0081] Based on the unified coordinate system and the obstacle map, use the improved A* path planning algorithm to generate the pressing path of the adiabatic module 4. This algorithm introduces a dynamic weight function on the basis of the traditional A* algorithm. The weight function considers three factors: path length, curvature, and distance to obstacles, and the weight ratios are 0.4, 0.3, and 0.3 respectively. The algorithm first constructs a grid in the three-dimensional space with a grid resolution of 10 mm, and then searches for the optimal path from the starting position (the current position of the robotic arm) to the target position (the installation position of the adiabatic module). To optimize the path curvature, the algorithm introduces a smoothing factor when calculating the node cost to ensure that the generated path meets the kinematic constraints of the robotic arm. The accuracy error of the finally generated pressing path is controlled within 1 mm.

[0082] In specific implementation, the specific steps to improve the A* path planning algorithm are as follows: First, based on the geometric model of the adiabatic module and the three-dimensional model of the installation reference plane, spatial coordinate alignment is performed through the point cloud registration algorithm to unify the coordinate systems of the adiabatic module model and the installation reference plane model, and the obstacle boundary data of the concave and convex regions on the inner wall of the concrete outer tank is extracted; Then, in the unified coordinate system, the working space is divided into three-dimensional grids, and each grid node represents a possible position and posture of the robotic arm. The grid resolution is set to 10mm to balance the computational complexity and path accuracy. Using the obstacle boundary data of the concave and convex regions on the inner wall of the concrete outer tank extracted by point cloud registration, an obstacle map is constructed to mark the areas that the robotic arm needs to avoid. A dynamic weight function is introduced, considering three factors: path length, path curvature smoothness, and distance from obstacles, with weight ratios of 0.4, 0.3, and 0.3 respectively. The weight function will be dynamically adjusted according to the current state of the robotic arm and the environment to adapt to the path planning requirements at different positions. Starting from the starting position of the adiabatic module (the current position of the robotic arm) as the starting point and the installation target position as the end point, heuristic function is used to guide the search. The heuristic function adopts a hybrid mode of Manhattan distance combined with Euclidean distance to estimate the minimum cost from the current node to the target node, and at the same time considers the joint motion range and speed constraints of the robotic arm to ensure search efficiency and path feasibility. During the search process, the path nodes are smoothed through cubic spline interpolation to optimize the path curvature radius and ensure the smooth movement of the end effector of the robotic arm; At the same time, it is verified whether the path meets the kinematic constraints of the robotic arm (such as joint angle range, speed, acceleration limit), and local replanning is performed on the path segments that do not meet the constraints. During the path planning process, multiple objectives such as the shortest path, the most labor-saving movement, and the optimal smoothness are comprehensively weighed, and a comprehensive score is assigned to each candidate path through the multi-attribute decision-making algorithm, and the path with the highest comprehensive score is selected as the final press-fitting path to ensure the efficient and precise operation of the robotic arm in a complex environment.

[0083] The press-fitting process is executed by an ABB IRB 6700 robotic arm, which has 6 degrees of freedom, a load capacity of 150kg, and a repeat positioning accuracy of ±0.1mm. An ATI Omega160 six-axis force sensor is installed at the end of the robotic arm, with a force measurement range of ±660N and a force resolution of 0.25N. The robotic arm moves the adiabatic module 4 to above the installation position according to the planned path, and then enters the precise alignment stage. In this stage, the system uses a method combining vision guidance and force control to identify the position of the expansion bolt 1 through the industrial camera installed on the robotic arm, and precisely adjusts the posture of the adiabatic module 4 to align the reserved hole with the bolt screw.

[0084] After the reserved holes are initially aligned with the bolts and screws, the robot arm switches to the force control mode, and adjusts the position and posture of the insulation module 4 through the real-time feedback of the six-dimensional force sensor to achieve precise alignment. After the alignment is completed, the robot arm applies vertical pressure, and the pressure value is set to 200N, which is evenly distributed on the surface of the insulation module 4. The pressure holding time is set to 30 seconds to ensure that the bottom surface of the insulation module is fully fitted with the top surface of the leveling gasket 3 to eliminate possible gaps. During the press-fitting process, the force sensor continuously monitors the pressure changes. If abnormal force feedback is detected, the system will automatically adjust the pressure or stop the operation to prevent damage to the insulation module 4.

[0085] S5: Install the fastening nut 2 on the screw portion of the expansion bolt 1;

[0086] Specifically, the fastening nut 2 is installed on the screw portion of the expansion bolt 1 by using a torque wrench, so that each fastening nut 2 is tightened to a preset torque value.

[0087] In this embodiment, after the insulation module 4 is press-fitted, it is necessary to install the fastening nut 2 on the screw portion of the expansion bolt 1 to ensure that the insulation module 4 is firmly fixed to the inner wall of the concrete outer tank. The fastening nut 2 is a nylon anti-loosening nut of M12 specification, made of 304 stainless steel, with good anti-loosening performance and corrosion resistance.

[0088] During the installation process, the operator uses a digital torque wrench for tightening operations. The measurement range of the digital torque wrench is 10-100N·m, and the accuracy is ±2%. According to the material characteristics and installation requirements of the insulation module 4, the preset tightening torque value is 45N·m. This torque value has been calculated and verified by experiments to provide sufficient tightening force without causing compression deformation to the insulation module 4.

[0089] The operator tightens the nuts from the center to the periphery in a predetermined order to ensure uniform force. During the tightening process, the torque wrench displays the current torque value in real time. When it reaches the preset value of 45N·m, the wrench will emit an audible and visual prompt, and the operator will stop rotating. After each nut is tightened, the system records the actual tightening torque value and tightening time to form installation quality traceability data.

[0090] To ensure long-term reliability, apply an appropriate amount of anti-loosening glue to the threaded part before tightening the nut. Use a medium-strength anti-loosening glue with a curing time of 24 hours. After complete curing, it can withstand an operating temperature of -40°C to 120°C, meeting the operating environment requirements of low-temperature storage tanks.

[0091] S6: Install the reference plane through 3D laser scanning to generate the actual flatness point cloud data, and conduct 3D deviation analysis with the 3D model of the installation reference plane obtained in S2. If the local error is greater than the preset threshold, use the adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance points, and perform iterative correction through the process of S3.

[0092] Specifically, use a 3D laser scanner to perform a full-section scan of the installation reference plane to generate the actual flatness point cloud data, and conduct 3D deviation analysis on the actual flatness point cloud data and the 3D model of the installation reference plane in S2. Use the ICP registration algorithm to align the model with the actual point cloud, calculate the local flatness error value, and mark it as an out-of-tolerance point if the error ≥ 2mm / m; use the adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance points, and perform iterative correction through the process of S3.

[0093] In this embodiment, the specific process of using the adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance points and performing iterative correction through the process of S3 includes: Subsequently, based on the coordinates of the out-of-tolerance points, use the adaptive leveling algorithm to recalculate the leveling compensation amount. The adaptive leveling algorithm iteratively optimizes the compensation amount ΔH by the gradient descent method, and the calculation formula is: ΔH = k·(H 实际 - H 理论 ), where k is the convergence coefficient, with a value range of 0.8 - 1.2, and feedback the corrected leveling parameter instruction set to the servo motor control process of S3, and drive the ball nut 6 to level again through the servo motor until the error is less than the preset threshold.

[0094] In this embodiment, after the adiabatic module 4 is installed, it is necessary to verify the installation quality. Use the FARO Focus S350 3D laser scanner to perform a full-section scan of the installation reference plane, and set the scanning accuracy to ±1mm. During the scanning process, set the scanner at a position that can cover the entire installation area and perform a 360-degree omnidirectional scan. To improve the scanning accuracy, increase the scanning density in the key areas, and the point cloud density reaches 2000 points / cm².

[0095] After the scanning is completed, process the original point cloud data through the FARO SCENE software to remove noise and generate the actual flatness point cloud data. Import this point cloud data into the CloudCompare software and conduct comparative analysis with the 3D model of the installation reference plane obtained in step S2. First, use the ICP registration algorithm to align the two models, and set the maximum number of iterations to 100 and the convergence threshold to 0.0005mm during the registration process. After the registration is completed, calculate the deviation value of each point to generate a deviation cloud map, intuitively showing the flatness error distribution of the installation reference plane.

[0096] The system automatically analyzes the deviation data and calculates the local flatness error value. The flatness error is calculated using the least squares method, and the maximum deviation from the fitted plane of the point cloud is calculated within a local area of 1m×1m. According to the insulation requirements of the cryogenic storage tank, the flatness error threshold is set to 2mm / m. The system automatically marks the areas with an error ≥2mm / m as out-of-tolerance points and records the coordinates and actual error values of these points.

[0097] For out-of-tolerance points, the system uses an adaptive leveling algorithm to recalculate the leveling compensation amount. This algorithm is based on the gradient descent method and determines the optimal compensation amount through iterative optimization. The calculation formula for the compensation amount ΔH is: ΔH = k·(H 实际 - H 理论 ), where H 实际 is the actual measured height, H 理论 is the theoretical design height, and k is the convergence coefficient. The value range of the convergence coefficient k is 0.8 - 1.2, and the initial value is set to 1.0. During the iteration process, the system dynamically adjusts the value of k according to the effect of the previous adjustment. If the adjustment effect is good, k is increased to accelerate convergence; if overshoot occurs, k is decreased to improve stability.

[0098] The calculated leveling compensation amount is integrated into the corrected leveling parameter instruction set and fed back to the servo motor control process in Step S3. The system automatically performs a secondary leveling process and precisely adjusts by driving the ball nut 6 with the servo motor. After the adjustment is completed, three-dimensional laser scanning verification is performed again. If the local error still exceeds the threshold, iterative correction continues until the error of all points is less than the preset threshold of 2mm / m.

[0099] Generally, after 1 - 2 iterations of correction, most out-of-tolerance points can meet the requirements. For individual points that are difficult to adjust, the system generates an exception report for manual inspection and processing by professional technicians. Finally, the system generates an installation quality report, including the flatness error distribution map, comparison data before and after adjustment, and the final acceptance result, providing a quality assurance basis for subsequent projects.

[0100] It should be noted that Example 1 is one of the installation construction methods for the insulation module of a cryogenic storage tank, and the above installation construction method can be applied to both LNG onshore tanks and LNG ships.

[0101] The above description of the embodiments is to enable those of ordinary skill in the art to understand and use the invention. It is obvious that those who are familiar with the technology in this field can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative efforts. Therefore, the present invention is not limited to the above embodiments, and all improvements and modifications made by those skilled in the art without departing from the scope of the present invention according to the disclosure of the present invention should be within the protection scope of the present invention.

Claims

1. A method for installing and constructing an insulation module of a low-temperature storage tank, characterized in that, It includes the following steps: S1. Drill holes at predetermined points on the inner wall of the concrete outer tank of the cryogenic storage tank and implant expansion bolts. Subsequently, obtain the surface topography data of the inner wall of the concrete outer tank of the cryogenic storage tank through three-dimensional laser scanning, and generate a point cloud model; S2. Based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, calculate the theoretical leveling height at each expansion bolt point, and generate a leveling parameter instruction set; S3. According to the leveling parameter instruction set, instruct the servo motor of the bolt tightening device to rotate the ball nut sleeved on the screw part of the expansion bolt, so that the leveling gasket moves axially along the screw to the target height; S4. Perform spatial matching between the geometric model of the insulation module and the three-dimensional model of the installation reference plane, generate a pressing path for the insulation module through a path planning algorithm, control the pressing robotic arm to align the reserved hole of the insulation module with the screw part of the expansion bolt, and press it with a preset vertical pressure so that the surface of the insulation module is fully attached to the surface of the leveling gasket; S5. Install a fastening nut on the screw part of the expansion bolt; S6. Generate actual flatness point cloud data by three-dimensional laser scanning the installation reference plane, perform three-dimensional deviation analysis with the three-dimensional model of the installation reference plane obtained in S2. If the local error is greater than the preset threshold, use an adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance point, and perform iterative correction through the process of S3.

2. The installation and construction method of a cryogenic storage tank insulation module according to claim 1, characterized in that In S1, it specifically includes the following steps: Drill holes at predetermined points on the inner wall of the concrete outer tank of the cryogenic storage tank and implant expansion bolts to form physical reference points for leveling and fixing; Scan the inner wall of the concrete outer tank after implanting the expansion bolts with a three-dimensional laser scanning device, obtain the surface topography data, and generate a point cloud model based on the data.

3. The installation and construction method of a cryogenic storage tank insulation module according to claim 1, characterized in that, In S2, it specifically includes the following steps: Based on the point cloud model, construct a three-dimensional model of the installation reference plane through the BIM system, and optimize the model stiffness distribution by using a finite element analysis algorithm; According to the three-dimensional model of the installation reference plane, calculate the theoretical leveling height at each expansion bolt point through a geometric feature extraction algorithm, and generate a leveling parameter instruction set including leveling height, rotation angle, and error threshold.

4. The installation and construction method of a cryogenic storage tank thermal insulation module according to claim 1, characterized in that, In S3, it specifically includes the following steps: Based on the leveling parameter instruction set, analyze the target height data of each point through a motion control algorithm, generate control instructions for the rotation angle, direction, and speed of the servo motor, where the accuracy of the control instructions matches the lead of the thread of the ball nut; Drive the servo motor of the bolt tightening device to rotate the ball nut according to the control instructions, and use a laser rangefinder to monitor the axial displacement of the leveling gasket in real time, and dynamically correct the rotation angle until the gasket moves to the target height.

5. The installation and construction method of a cryogenic storage tank insulation module according to claim 1, characterized in that, In S4, it specifically includes the following steps: Coat resin mastic on the back of the insulation module; Based on the geometric model of the insulation module and the three-dimensional model of the installation reference plane, perform spatial matching through an improved A* path planning algorithm to generate an obstacle-avoiding optimized pressing path; Control the press-fitting robotic arm to move along the press-fitting path, and real-time feedback of pose data is obtained through a six-axis force sensor. Align the reserved hole of the adiabatic module precisely with the screw part of the expansion bolt, and apply a vertical pressure. The pressure holding time is ≥ 30 seconds to make the bottom surface of the adiabatic module form a full-contact surface with the top surface of the leveling gasket; At the same time, make the resin mastic adhesively connect the press-fitted adiabatic module to the concrete exterior wall.

6. The installation and construction method of a cryogenic storage tank insulation module according to claim 5, characterized in that, In S4, the specific process of generating the obstacle-avoiding optimized press-fitting path includes: Based on the geometric model of the adiabatic module and the three-dimensional model of the installation reference plane, spatial coordinate alignment is performed through a point cloud registration algorithm to unify the coordinate systems of the adiabatic module model and the installation reference plane model, and the obstacle boundary data of the concave and convex areas on the inner wall of the concrete outer tank are extracted; In the unified coordinate system, a press-fitting path is generated based on the improved A* path planning algorithm. In the improved A* path planning algorithm, the path curvature is optimized through a dynamic weight function to avoid the obstacle boundary and meet the kinematic constraints of the robotic arm. The path planning accuracy error is ≤ 1 mm.

7. A method for installing a cryogenic storage tank insulation module according to claim 1, characterized in that, In S5, it specifically includes the following steps: Install a fastening nut on the screw part of the expansion bolt through a torque wrench so that each fastening nut is fastened to the preset torque value.

8. The installation and construction method of a cryogenic storage tank insulation module according to claim 1, characterized in that, In S6, it specifically includes the following steps: The installation reference plane is scanned in full section by a three-dimensional laser scanner to generate actual flatness point cloud data, and three-dimensional deviation analysis is performed on the actual flatness point cloud data and the three-dimensional model of the installation reference plane in S2. The ICP registration algorithm is used to align the model with the actual point cloud, and the local flatness error value is calculated. If the error ≥ 2 mm / m, it is marked as an out-of-tolerance point position; The adaptive leveling algorithm is used to recalculate the leveling compensation amount of the out-of-tolerance point position, and iterative correction is performed through the process of S3.

9. The installation and construction method of a cryogenic storage tank insulation module according to claim 8, characterized in that, The specific process of using the adaptive leveling algorithm to recalculate the leveling compensation amount of the out-of-tolerance point position and performing iterative correction through the process of S3 includes: Subsequently, based on the coordinates of the out-of-tolerance point position, the adaptive leveling algorithm is used to recalculate the leveling compensation amount. The adaptive leveling algorithm iteratively optimizes the compensation amount ΔH by the gradient descent method, and the calculation formula is: ΔH = k·(H 实际 - H 理论 ), where H 实际 is the actually measured height, H 理论 is the theoretically designed height, k is the convergence coefficient, and the value range of k is 0.8 - 1.2 And the corrected leveling parameter instruction set is fed back to the servo motor control process of S3, and the ball nut is driven by the servo motor for secondary leveling until the error is less than the preset threshold.

10. A method for installing and constructing an insulation module of a low-temperature storage tank according to claim 1, characterized in that, In S3, the output end of the servo motor of the bolt tightening device can be connected to the ball nut through a rotating kit to achieve torque transmission.

Images