A Weld Detection Method, Medium and System for the Welding of Floating Ship Plates of Oil Storage Tanks

Abstract

The present invention provides a weld detection method, medium and system for the welding of floating deck plates of oil storage tanks, belonging to the technical field of electronic digital data processing, and realizing comprehensive and accurate monitoring through six key matrices. Starting from the floating deck plate motion feature matrix, height, attitude and weld displacement data are accurately captured. Through multi-source detection technology, a comprehensive detection matrix including ultrasonic flaw detection, X-ray imaging and surface topography is established to comprehensively characterize the microscopic features of the weld. In-depth dynamic stress and strain analysis is carried out to construct a complex stress field distribution model considering the floating effect. The multi-dimensional feature matrices are combined by weights to form a comprehensive weld quality evaluation matrix, and an adaptive detection parameter update mechanism is established. Finally, a weld state vector and a safety threshold matrix are constructed to realize real-time monitoring and early warning, solving the technical problem that the prior art usually can only statically evaluate the internal quality of the weld and cannot comprehensively reflect the actual stress and strain state of the weld under dynamic working conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic digital data processing. Specifically, it relates to a method, medium, and system for detecting weld seams in the welding of floating plates of oil storage tanks. Background Art

[0002] Floating storage tanks are widely used in the fields of offshore oil exploitation and storage, and are one of the key equipment to ensure the efficient utilization of marine oil resources. Such floating storage tanks usually consist of a cylindrical floating plate and an oil storage tank located inside it. The function of the floating plate is to provide buoyancy support for the oil storage tank, ensuring that the entire storage tank system can float stably on the sea surface. Compared with fixed storage tanks, floating storage tanks have higher adaptability and flexibility, and can cope with complex marine environmental conditions.

[0003] However, during the long-term operation of floating storage tanks at sea, their structures are severely impacted by external forces such as waves, wind waves, and ocean currents, resulting in various deformations and damages to the floating plates, ultimately affecting the service life and safety of the entire storage tank. The welded joint parts on the floating plates are the key parts most prone to problems. Due to the action of complex external dynamic loads, these weld seams are prone to failures such as fractures and cracks, and may even cause serious environmental pollution accidents. Therefore, how to effectively detect and evaluate the quality status of the weld seams of floating storage tanks and timely discover abnormal situations has become an urgent technical problem to be solved.

[0004] In response to the above problems, some existing detection methods have been proposed in the industry, such as using ultrasonic flaw detection, X-ray imaging, etc. to conduct conventional non-destructive testing on weld seams. However, these methods usually can only statically evaluate the internal quality of weld seams and cannot comprehensively reflect the actual stress and strain states of weld seams under dynamic working conditions. In addition, the existing detection methods also have problems such as unreasonable parameter settings, low detection accuracy, and inability to achieve real-time monitoring. Therefore, there is an urgent need for a new type of weld seam detection method that can comprehensively analyze the dynamic mechanical behavior of weld seams and achieve adaptive parameter optimization and real-time warning functions, thereby improving the safety and reliability of the entire storage tank system. Summary of the Invention

[0005] In view of this, the present invention provides a method, medium, and system for detecting weld seams in the welding of floating plates of oil storage tanks, which can solve the technical problem that the existing technology usually can only statically evaluate the internal quality of weld seams and cannot comprehensively reflect the actual stress and strain states of weld seams under dynamic working conditions.

[0006] The present invention is implemented as follows:

[0007] The first aspect of the present invention provides a method for detecting weld seams in the welding of floating plates of oil storage tanks, including the following steps:

[0008] S10. Construct a motion feature matrix of the floating pontoon board, including: collecting the height matrix of the floating pontoon board and recording the real-time height data of measuring points at different positions; establishing an attitude matrix of the floating pontoon board, including pitch angle, roll angle, yaw angle and velocity components in three directions; obtaining the relative displacement matrix of the circumferential weld and the radial weld during the up-and-down floating process of the floating pontoon board;

[0009] S20. Collect a weld basic detection matrix, including: establishing an ultrasonic flaw detection feature matrix and recording the ultrasonic reflection coefficients at different positions and different detection angles; constructing an X-ray imaging feature matrix and obtaining the X-ray penetration rates at different positions and different exposure parameters; forming a weld surface topography feature matrix, including weld contour and surface defect distribution information;

[0010] S30. Conduct dynamic stress and strain analysis: establish a stress matrix, including normal stress and shear stress components; construct a strain matrix, including normal strain and shear strain components; calculate the stress field distribution considering the floating influence through the stress-strain correlation equations;

[0011] S40. Construct a comprehensive weld quality evaluation matrix: perform weighted combination of the ultrasonic flaw detection feature matrix, the X-ray imaging feature matrix and the stress-strain matrix; establish a weight coefficient matrix to ensure reasonable proportion of each index; form an evaluation matrix reflecting the overall quality state of the weld;

[0012] S50. Realize adaptive adjustment of detection parameters: establish a detection parameter matrix including ultrasonic angle, X-ray parameters and strain acquisition parameters; construct a parameter update equation based on the weld quality evaluation matrix; realize dynamic optimization update of detection parameters;

[0013] S60. Execute real-time monitoring and early warning: construct a weld state vector, including integrity index, stress concentration index and fatigue damage index; set a safety threshold matrix and establish early warning trigger conditions; realize real-time monitoring of weld state and abnormal early warning.

[0014] Among them, the stress-strain correlation equations include a principal stress calculation equation, a shear stress calculation equation, a strain distribution equation and a fatigue damage accumulation equation;

[0015] The principal stress calculation equation is used to calculate the magnitude and direction of the principal stress at each point of the weld. The inputs include the weld position coordinates, the floating pontoon board height data, the attitude angle data and the velocity components, and the outputs are the magnitude and direction of the principal stress at each measuring point;

[0016] The shear stress calculation equation is used to determine the shear stress distribution borne by the weld. The inputs include the principal stress data, the weld relative displacement data and the material shear modulus, and the outputs are the shear stress components at each point of the weld and the position of the maximum shear stress;

[0017] The strain distribution equation is used to calculate the deformation state of the weld. The inputs include principal stress data, shear stress data, material elastic modulus, and Poisson's ratio, and the outputs are the normal strain and shear strain distributions in the weld area.

[0018] The fatigue damage accumulation equation is used to evaluate the fatigue state of the weld. The inputs include the number of stress cycles, stress amplitude, mean stress, and material fatigue parameters, and the output is the fatigue damage degree at each point of the weld.

[0019] The parameter update equation is used to optimize the detection parameter settings. The inputs include the current detection parameters, weld quality evaluation index, detection accuracy requirements, and equipment performance constraints, and the outputs are the updated ultrasonic incident angle, X-ray exposure parameters, and strain acquisition frequency.

[0020] Among them, the calculation process involved in the present invention is described in detail as follows:

[0021] 1. The motion characteristic matrix of the pontoon plate is expressed as follows:

[0022] ;

[0023] In the formula, represents the height value of the th measurement point at the th moment, with the unit of meter, is the number of measurement points, is the number of sampling moments;

[0024] The attitude matrix is expressed as follows:

[0025] ;

[0026] In the formula, is the pitch angle, with the range of ; is the roll angle, with the range of ; is the yaw angle, with the range of ; are the velocity components in three directions, with the unit of m / s;

[0027] The relative displacement matrix is expressed as follows:

[0028] ;

[0029] In the formula, is the circumferential weld relative displacement vector, is the radial weld relative displacement vector;

[0030] 2. The principal stress calculation equation is expressed as follows:

[0031] ;

[0032] In the formula, is the stress in the x direction, is the stress in the y direction, is the shear stress, is the density of the oil, is the acceleration due to gravity, is the liquid level height, is the attitude influence coefficient, is the dynamic viscosity; represents the principal stress: Here, the subscripts 1 and 2 represent two principal stress values, is the maximum principal stress, is the minimum principal stress. Stress is the normal stress corresponding to zero shear stress when the coordinate system is rotated to the appropriate direction at a specific point.

[0033] The first term is the average normal stress, reflecting the average stress level in two principal directions;

[0034] The second term is the maximum shear stress, characterizing the degree of shear deformation;

[0035] Among them:

[0036] In the term, is the hydrostatic pressure term, is the attitude correction term;

[0037] The structure is the same as above, but the coefficients are different to reflect anisotropy;

[0038] is the shear stress caused by velocity;

[0039] 3. The shear stress calculation equation is expressed as follows:

[0040] ;

[0041] In the formula, is the shear modulus, is the shear strain, is the viscosity coefficient, is the second derivative term coefficient;

[0042] The first term is the static shear stress, reflecting the elastic deformation of the material;

[0043] The second term is the viscous stress, describing the influence of the strain rate;

[0044] The third term and the fourth term is the second - order spatial derivative term, reflecting the propagation effect of strain;

[0045] 4. The strain distribution equation is expressed as follows:

[0046] ;

[0047] In the formula, is the elastic modulus, is the Poisson's ratio, ; are the components of the stress tensor, is the subscript indicating the stress direction, taking values of 1, 2, 3. The first subscript i represents the normal direction of the plane where the stress is located, and the second subscript j represents the stress direction. For example, represents the stress component along the 2 - direction on the plane perpendicular to the 1 - direction; represents the trace of the stress tensor: The repeated appearance of k indicates summation over k = 1, 2, 3: ; It represents the sum of the normal stress components; is the Kronecker symbol. When i = j, When i≠j, , which is used to represent the tensor operation in the strain equation;

[0048] The first term is the normal strain component;

[0049] The second term is the transverse deformation coupling term;

[0050] and The relationship between: is the eigenvalue calculated through ; The principal stress calculation equation converts ( ), ( ) and ( ) into the principal stress . This conversion actually finds the principal axis direction of the stress state, such that there is only normal stress and no shear stress in this direction; In the plane stress state, the principal stress calculation formula is the one given in the text:

[0051] ;

[0052] The physical meaning of this transformation is that for the stress state at any point, a specific coordinate system can always be found such that in this coordinate system, the point only bears normal stress (tensile or compressive) without shear action. The direction of this coordinate system is the principal stress direction, and the corresponding normal stress value is the principal stress value.

[0053] 5. The fatigue damage accumulation equation is expressed as follows:

[0054] ;

[0055] In the formula, is the actual number of cycles, is the fatigue life, is the damage index, is the correction coefficient, is the stress amplitude, is the fatigue limit, is the material constant;

[0056] The first term is the damage accumulation term of the number of cycles;

[0057] The second term is the damage accumulation term of the stress amplitude;

[0058] 6. The parameter update equation is expressed as follows:

[0059] ;

[0060] In the formula, is the current parameter vector, is the learning rate, is the performance index function, is the second derivative weight, is the random perturbation term.

[0061] The first term is the current parameter value;

[0062] The second term is the first-order gradient correction term;

[0063] The third term is the second-order gradient correction term;

[0064] The fourth term is the random perturbation term.

[0065] The method for obtaining the parameter is as follows:

[0066] 1. It is obtained by measuring with an oil density meter at a measurement temperature of , and the number of measurements is not less than 3 times, and the average value is taken;

[0067] 2. Adopt the local gravitational acceleration value, which can be obtained by measuring with a gravimeter;

[0068] 3. Obtained by real-time measurement with a liquid level gauge, and the sampling frequency is not less than ;

[0069] 4. Obtained by finite element analysis and fitting of experimental data:

[0070] Step 1: Establish a finite element model of the floating plate;

[0071] Step 2: Apply different attitude angle conditions;

[0072] Step 3: Extract stress data;

[0073] Step 4: Obtain the coefficient by fitting with the least square method;

[0074] 5. Obtained by measuring with a rotational viscometer, and the measurement temperature range is , at intervals of ;

[0075] 6. Obtained by a material tensile test:

[0076] Step 1: Prepare standard specimens;

[0077] Step 2: Conduct a tensile test;

[0078] Step 3: Calculate according to the stress-strain curve;

[0079] 7. Obtained by measuring with a dynamic mechanical analyzer, and the test frequency range is ;

[0080] 8. Obtained by a wave propagation experiment:

[0081] Step 1: Apply a transient load;

[0082] Step 2: Measure the strain propagation speed;

[0083] Step 3: Inversely solve to obtain the coefficient;

[0084] 9. Obtained by standard material mechanics performance tests;

[0085] 10. Obtained by a fatigue test:

[0086] Step 1: Conduct a constant amplitude fatigue test;

[0087] Step 2: Record the number of fracture cycles;

[0088] Step 3: Determine the exponent using regression analysis;

[0089] 11. Determine through comparative tests:

[0090] Step 1: Conduct variable amplitude fatigue tests;

[0091] Step 2: Measure the actual life;

[0092] Step 3: Determine the correction factor by comparing with theoretical calculations;

[0093] 12. Determine through optimizing algorithm performance testing:

[0094] Step 1: Set the initial values;

[0095] Step 2: Conduct convergence testing;

[0096] Step 3: Select the optimal parameter combination.

[0097] Explanation of the above equations:

[0098] 1. The principal stress calculation equation uses quadratic terms to describe the influence of the attitude angle on stress, considering the unique motion characteristics of the floating pontoon plate, because the influence of attitude changes on stress has non-linear characteristics;

[0099] 2. The shear stress calculation equation introduces second-order derivative terms of time and space, which takes into account the propagation characteristics of strain and boundary effects;

[0100] 3. The strain distribution equation is based on Hooke's law in general form, considering the coupling effect under three-dimensional stress states;

[0101] 4. The fatigue damage accumulation equation uses a power function relationship, which is in line with the non-linear cumulative characteristics of material fatigue damage;

[0102] 5. The parameter update equation combines the characteristics of gradient descent and Newton's method, ensuring both the convergence speed and improving the optimization accuracy.

[0103] The second aspect of the present invention provides a computer-readable storage medium, wherein program instructions are stored in the computer-readable storage medium, and when the program instructions run on a computer, they are used to execute the above-mentioned weld detection method for welding of floating pontoon plates of oil storage tanks.

[0104] The third aspect of the present invention provides a weld detection system for welding of floating pontoon plates of oil storage tanks, which includes the above-mentioned computer-readable storage medium.

[0105] Compared with the prior art, the beneficial effects of a weld detection method, medium and system for welding the floating deck of an oil storage tank provided by the present invention are as follows:

[0106] First, the method constructs a motion feature matrix of the floating deck, including a height matrix, an attitude matrix, and a relative displacement matrix, which comprehensively describes the dynamic deformation characteristics of the floating deck under complex sea conditions. This provides the necessary basic data for subsequent stress and strain analysis, greatly improving the accuracy and reliability of the detection results.

[0107] Second, the method uses matrix operations to replace the traditional data processing flow. This not only improves the calculation efficiency but also facilitates the comprehensive integration of the results of various detection means. At the same time, the method establishes a stress-strain correlation equation set to achieve precise analysis of the dynamic mechanical behavior of the weld, laying a solid theoretical foundation for the comprehensive evaluation of weld quality.

[0108] In addition, the method realizes the adaptive adjustment of detection parameters. By dynamically optimizing key parameters such as ultrasonic angle, X-ray exposure, and strain acquisition frequency, it ensures the best performance of various detection means and greatly improves the detection accuracy.

[0109] Finally, the method has real-time monitoring and early warning functions. It can continuously monitor key indicators such as weld integrity, stress concentration, and fatigue damage, and immediately issue an early warning once an abnormal situation is detected. This greatly improves the safety and reliability of the entire storage tank system and provides a strong guarantee for the efficient utilization of offshore oil resources.

[0110] Generally speaking, the weld detection method of the present invention fully considers the unique dynamic working conditions of floating storage tanks and uses advanced matrix analysis and parameter optimization technologies to achieve comprehensive evaluation and real-time monitoring of weld quality. It has made remarkable technological progress in improving detection accuracy and safety, and solves the technical problem that the prior art usually can only statically evaluate the internal quality of welds and cannot fully reflect the actual stress and strain state of welds under dynamic working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 It is a flowchart of the method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0112] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0113] As Figure 1 shown, it is a flowchart of a weld detection method for welding the floating deck of an oil storage tank provided by the present invention. The following is a detailed description of the specific implementation of each step:

[0114] First, for step S10, it is necessary to construct the motion feature matrix of the floating pontoon plate. It includes:

[0115] Height matrix :

[0116] ;

[0117] Among them, represents the height value of the th measurement point at the th moment, with the unit of meter; is the number of measurement points, is the number of sampling moments. By analyzing this height matrix, the displacement change of the floating pontoon plate in the vertical direction can be understood.

[0118] Attitude matrix :

[0119] ;

[0120] Among them, is the pitch angle, with the range of ; is the roll angle, with the range of ; is the yaw angle, with the range of ; are the velocity components in three directions respectively, with the unit of m / s. This matrix reflects the attitude change and motion characteristics of the floating pontoon plate in the horizontal and vertical directions.

[0121] Relative displacement matrix :

[0122] ;

[0123] Among them, is the relative displacement vector of the circumferential weld, is the relative displacement vector of the radial weld. By analyzing these relative displacement data, the deformation of the weld in different directions can be understood.

[0124] Generally speaking, step S10 constructs the motion feature matrix of the floating pontoon plate, providing the necessary basic data for subsequent weld detection.

[0125] Next, for step S20, it is necessary to collect the basic detection matrix of the weld. It includes:

[0126] Ultrasonic flaw detection feature matrix:

[0127] This matrix records the ultrasonic reflection coefficient data at different positions and different detection angles. By analyzing these data, the defects and discontinuities inside the weld can be understood.

[0128] X-ray imaging feature matrix:

[0129] This matrix obtains the X-ray penetration rate data at different positions and different exposure parameters. By analyzing these data, the overall quality of the weld, including internal defects, pores, etc., can be understood.

[0130] Weld surface topography feature matrix:

[0131] This matrix contains the contour information of the weld and the distribution of surface defects. By analyzing these data, the appearance characteristics of the weld can be understood, providing a basis for subsequent comprehensive evaluation.

[0132] By collecting the weld feature matrices obtained by these three different detection methods, the information on the internal quality, surface quality, and deformation characteristics of the weld can be comprehensively understood, providing the necessary basic data for subsequent stress-strain analysis and comprehensive evaluation.

[0133] Next, for step S30, dynamic stress-strain analysis needs to be carried out. It includes:

[0134] Construction of the stress matrix:

[0135] The stress matrix contains normal stress components and shear stress components. These stress data are obtained from the height matrix , attitude matrix and relative displacement matrix obtained in step S10 through the principal stress calculation equation:

[0136] ;

[0137] Among them, is the stress in the x direction, is the stress in the y direction, is the shear stress. is the oil density, is the acceleration due to gravity, is the liquid level height, is the attitude influence coefficient, is the dynamic viscosity. This equation takes into account the hydrostatic pressure, attitude angle, and motion speed of the floating ship plate's influence on stress and can accurately reflect the stress state borne by the weld.

[0138] Construction of the strain matrix:

[0139] The strain matrix contains normal strain components and shear strain components. These strain data are calculated through the strain distribution equation based on the stress matrix obtained above and the material mechanical property parameters:

[0140] ;

[0141] where, is the elastic modulus, is the Poisson's ratio, is the Kronecker symbol, . This equation is based on the generalized Hooke's law and considers the coupling effect of strains under three-dimensional stress states.

[0142] Application of the stress-strain correlation equations:

[0143] Furthermore, with the help of the stress-strain correlation equations, the stress field distribution considering the floating effect is calculated. It includes:

[0144] Principal stress calculation equation: same as above

[0145] Shear stress calculation equation: ;

[0146] Strain distribution equation: same as above

[0147] Fatigue damage accumulation equation: ;

[0148] These equations fully consider the unique motion characteristics of the floating ship plate and provide an important basis for subsequent comprehensive evaluation.

[0149] Generally speaking, step S30 realizes the accurate analysis of the dynamic stress-strain state of the weld by establishing the stress matrix and strain matrix and combining the stress-strain correlation equations.

[0150] Next, for step S40, a comprehensive evaluation matrix of weld quality needs to be constructed. It includes:

[0151] Weighted combination of feature matrices:

[0152] The ultrasonic flaw detection feature matrix, X-ray imaging feature matrix obtained in step S20, and the stress-strain matrix obtained in steps S301 and S302 are weighted and combined to reflect the importance degree of different detection means for weld quality evaluation:

[0153] ;

[0154] where, represent the feature matrices of ultrasonic flaw detection, X-ray imaging, and stress-strain respectively, are the corresponding weight coefficients, satisfying .

[0155] Determination of the weight coefficient matrix:

[0156] Establish the weight coefficient matrix , ensuring a reasonable proportion of each index in the comprehensive evaluation. For example, it can be set , so that ultrasonic flaw detection accounts for 40%, X-ray imaging accounts for 30%, and stress and strain account for 30%. By adjusting these weight coefficients, the weight distribution of the evaluation indexes can be flexibly determined according to different application scenarios and detection requirements.

[0157] Formation of the comprehensive evaluation matrix:

[0158] Finally, form an evaluation matrix reflecting the overall quality status of the weld . This matrix comprehensively considers various factors such as internal defects, appearance quality, and mechanical properties of the weld, and gives an overall quality evaluation result.

[0159] Generally speaking, step S40 constructs a comprehensive evaluation matrix. By weighted combination of the characteristic data of different detection means, an evaluation result comprehensively reflecting the quality status of the weld is obtained. This lays a foundation for subsequent adaptive adjustment and real-time monitoring.

[0160] Next, for step S50, it is necessary to realize the adaptive adjustment of the detection parameters. These include:

[0161] Establishment of the detection parameter matrix:

[0162] Establish a detection parameter matrix including ultrasonic angle, X-ray parameters, and strain acquisition parameters . These parameters determine the measurement accuracy and effectiveness of various detection means.

[0163] Construction of the parameter update equation:

[0164] Based on the weld quality evaluation matrix obtained in step S40 , construct the following parameter update equation:

[0165] ;

[0166] Wherein, is the current parameter vector, is the learning rate, is the performance index function (i.e., ), is the second derivative weight, is the random perturbation term. This equation combines the characteristics of the gradient descent method and the Newton method, ensuring both the convergence speed and the optimization accuracy. By continuously iterating and updating, a set of optimal detection parameter settings can be found to ensure the accuracy of the evaluation result.

[0167] Explanation of the parameter update process:

[0168] The parameter update equation consists of four terms:

[0169] The first term is the current parameter value;

[0170] The second term is the first-order gradient correction term, which is used to adjust the direction of the parameter;

[0171] The third term is the second-order gradient correction term, which is used to accelerate the convergence speed;

[0172] The fourth term is the random perturbation term, which is beneficial to jumping out of the local optimal solution.

[0173] Generally speaking, step S50 realizes the adaptive adjustment of the detection parameters, makes full use of the weld quality evaluation results to dynamically optimize various detection parameters, and improves the performance and accuracy of the entire detection system.

[0174] Finally, for step S60, real-time monitoring and early warning need to be performed. It includes:

[0175] Construction of the weld state vector:

[0176] Construct the weld state vector , including the integrity index , the stress concentration index and the fatigue damage index . These indexes reflect the overall quality status of the weld and provide a basis for subsequent safety early warning.

[0177] Setting of the safety threshold matrix:

[0178] Set the safety threshold matrix , and set reasonable safety upper limit values for each index. For example, . Once any index exceeds the threshold, the early warning mechanism will be triggered.

[0179] Real-time monitoring and abnormal early warning:

[0180] The system will continuously collect and analyze the data of each index , once any index is found, an early warning signal will be immediately sent out. In this way, abnormal weld quality can be detected in time, providing support for subsequent maintenance and improving the overall safety.

[0181] Generally speaking, step S60 establishes a set of real-time monitoring and early warning mechanisms. By setting reasonable safety thresholds, key indicators such as the integrity of the weld, stress concentration, and fatigue damage of the weld are monitored in real time. Once an abnormal situation is detected, an early warning can be issued in a timely manner, providing support for subsequent maintenance and ensuring the safe operation of the system.

[0182] The second aspect of the present invention provides a computer-readable storage medium, wherein program instructions are stored in the computer-readable storage medium, and when the program instructions run on a computer, they are used to execute the above-mentioned weld detection method for the floating deck welding of an oil storage tank.

[0183] The third aspect of the present invention provides a weld detection system for the floating deck welding of an oil storage tank, which includes the above-mentioned computer-readable storage medium.

[0184] Specifically, the principle of the present invention is as follows: By making full use of the motion characteristic data of the floating deck and through matrix operations, the accurate analysis of the dynamic stress and strain state of the weld is realized, and based on this, comprehensive quality evaluation and real-time monitoring and early warning are carried out. The key to this method lies in establishing a close connection between the motion characteristics of the floating deck and the mechanical behavior of the weld.

[0185] First, a motion characteristic matrix of the floating deck is constructed, including a height matrix, an attitude matrix, and a relative displacement matrix. Among them, the height matrix records the real-time height data of each measuring point at different times, reflecting the displacement change of the floating deck in the vertical direction; the attitude matrix includes pitch angle, roll angle, yaw angle, and velocity components in three directions, describing the attitude change of the floating deck in the horizontal and vertical directions; the relative displacement matrix gives the deformation of the circumferential weld and the radial weld during the floating process. These data lay a necessary foundation for subsequent stress and strain analysis.

[0186] Secondly, three common non-destructive testing characteristics are collected for the weld, namely ultrasonic flaw detection, X-ray imaging, and surface topography. These test results are recorded in the form of a matrix, providing multi-angle data support for the comprehensive evaluation of the weld quality.

[0187] On this basis, a stress-strain correlation equation set is established, including equations for principal stress calculation, shear stress calculation, strain distribution, and fatigue damage accumulation. These equations fully consider the unique dynamic working conditions of the floating deck, such as hydrostatic pressure, attitude angle change, motion speed, etc., and accurately simulate the stress and strain state borne by the weld. Through matrix operations, the stress and strain distribution of each part of the weld can be quickly obtained, providing a theoretical basis for subsequent comprehensive evaluation.

[0188] To ensure the accuracy of the detection results, the adaptive adjustment of detection parameters is also realized. Specifically, an optimization matrix including key parameters such as ultrasonic angle, X-ray parameters, and strain acquisition frequency is established. Using the weld quality evaluation result as the optimization goal, through the parameter update equation combining gradient descent and Newton's method, the optimal combination of detection parameters can be dynamically found, continuously improving the performance of the entire detection system.

[0189] Finally, a weld state monitoring and early warning mechanism is constructed. Three indicators reflecting the overall quality of the weld are defined, namely the integrity indicator, the stress concentration indicator, and the fatigue damage indicator, and corresponding safety thresholds are set. The system will collect and analyze the data of these indicators in real time. Once any indicator exceeds the threshold, an early warning signal will be immediately issued. This can not only timely detect the abnormal weld quality, but also provide strong support for subsequent maintenance, thus ensuring the safe operation of the entire storage tank system.

[0190] The following provides an embodiment of a specific application scenario of the present invention: An offshore oil enterprise conducts floating storage tank operation business in a certain area of the South China Sea. The storage tank system consists of a cylindrical floating ship plate with a diameter of 50 meters and a height of 30 meters and an internal oil storage tank of 100,000 cubic meters. To ensure the safe operation of the storage tank system, the enterprise decides to adopt the weld detection method proposed by the present invention to comprehensively evaluate and real-time monitor the welds on the floating ship plate.

[0191] First, the enterprise organizes relevant technical personnel to measure and model the motion characteristics of the floating ship plate according to step S10 of the present invention.

[0192] Specifically, the technical personnel set 100 height measurement points on the surface of the floating ship plate and use hydrophones to record the height data of each measurement point in real time. After 24 hours of continuous acquisition, a height matrix H of 100 rows and 100 columns is formed:

[0193] ;

[0194] Among them, each element represents the height value of the i-th measurement point at the j-th moment, with the unit of meter. By analyzing this height matrix, the displacement change of the floating ship plate in the vertical direction can be understood.

[0195] In addition, the enterprise sets a sensor at each of the three angles and three directions of the floating ship plate to real-time monitor the attitude angle and motion speed of the floating ship plate. After 24 hours of continuous acquisition, an attitude matrix A is formed:

[0196] ;

[0197] Among them, each row of data respectively represents the pitch angle roll angle Yaw angle and the velocity components in the x, y, and z directions . These data reflect the attitude changes and motion characteristics of the floating plate in the horizontal and vertical directions.

[0198] In addition, the enterprise also installed displacement sensors on the surface of the floating plate to monitor the relative displacement of the circumferential weld and the radial weld in real time. After 24 hours of continuous acquisition, a relative displacement matrix D was formed:

[0199] ;

[0200] where, m represents the relative displacement of the circumferential weld, m represents the relative displacement of the radial weld.

[0201] Through the above measurements and modeling, the enterprise has mastered the motion characteristic data of the floating plate under complex sea conditions, laying a foundation for subsequent weld detection.

[0202] Next, the enterprise collects the basic detection characteristics of the weld according to step S20 of the present invention.

[0203] First, the technician uses an ultrasonic flaw detector to scan and detect the weld on the floating plate. By adjusting the probe angle and position, the technician obtains an ultrasonic reflection coefficient matrix of 100 rows and 100 columns :

[0204] ;

[0205] where each element represents the ultrasonic reflection coefficient at the i-th detection position and the j-th detection angle. By analyzing these data, the defects and discontinuities inside the weld can be understood.

[0206] In addition, the enterprise also uses X-ray imaging technology to comprehensively scan the weld of the floating plate. After scanning with multiple different exposure parameters, the technician obtains an X-ray penetration rate matrix of 100 rows and 100 columns :

[0207] ;

[0208] where each element represents the X-ray penetration rate at the i-th detection position and the j-th exposure parameter. By analyzing these data, the overall quality of the weld, including internal defects, pores, etc., can be understood.

[0209] Finally, the enterprise uses a 3D scanner to collect the surface topography of the floating ship plate weld. After data processing, the technicians obtain a weld contour and defect distribution matrix with 100 rows and 100 columns. :

[0210] ;

[0211] Among them, each element represents the weld contour height at the i-th detection position and the defect depth at the j-th detection position. By analyzing these data, the appearance quality characteristics of the weld can be understood.

[0212] To sum up, through actual measurement and data collection, the enterprise constructs the motion characteristic matrix of the floating ship plate and the basic detection matrix of the weld, laying a foundation for subsequent stress-strain analysis and comprehensive evaluation.

[0213] Subsequently, the enterprise deeply analyzes the dynamic stress-strain state of the floating ship plate weld according to step S30 of the present invention.

[0214] First, based on the previously obtained height matrix H, attitude matrix A, and relative displacement matrix D, the enterprise calculates the stress matrix of the weld :

[0215] ;

[0216] Among them, each element in the matrix represents the magnitude of the principal stress at the i-th weld position, with the unit of megapascal. By analyzing this stress matrix, the stress state of the weld under complex sea conditions can be clearly understood.

[0217] On this basis, the enterprise also calculates the strain matrix of the weld according to the material mechanical property parameters :

[0218] ;

[0219] Among them, each element in the matrix represents the equivalent strain value at the i-th weld position, dimensionless. By analyzing this strain matrix, the deformation of the weld under the action of force can be understood.

[0220] Next, the enterprise further analyzed the dynamic mechanical behavior of the weld seam by using the stress-strain correlation equations. For example, according to the principal stress calculation equation, the technicians obtained the magnitudes and directions of the principal stresses at various parts of the weld seam; according to the shear stress calculation equation, the technicians determined the shear stress distribution borne by the weld seam; according to the strain distribution equation, the technicians calculated the normal strain and shear strain states in the weld seam area; according to the fatigue damage accumulation equation, the technicians evaluated the degree of fatigue damage of the weld seam. These results provided a theoretical basis for the subsequent comprehensive evaluation.

[0221] Based on the above analysis, the enterprise mastered the complete stress-strain state information of the floating ship plate weld seam under complex sea conditions, laying a foundation for the subsequent quality evaluation.

[0222] Immediately afterwards, the enterprise constructed a comprehensive evaluation matrix for weld quality according to step S40 of the present invention.

[0223] First of all, the enterprise combined the ultrasonic flaw detection matrix obtained above, the X-ray imaging matrix and the stress-strain matrix through weighted combination to form a comprehensive evaluation matrix for weld quality

[0224] ;

[0225] Among them, the weights of each detection method are set to 0.4, 0.3 and 0.3 respectively to reflect their importance in the evaluation of weld quality.

[0226] By analyzing this comprehensive evaluation matrix , the enterprise can comprehensively understand the overall quality state of the floating ship plate weld seam, including various indexes such as internal defects, appearance quality and mechanical properties. This provides a key basis for the subsequent optimization of detection parameters and safety monitoring.

[0227] Next, the enterprise adaptively adjusted the detection parameters according to step S50 of the present invention.

[0228] First of all, the enterprise established a detection parameter matrix including the ultrasonic flaw detection angle, X-ray exposure parameters and strain acquisition frequency

[0229] ;

[0230] Among them, each row corresponds to a set of detection parameter settings.

[0231] Based on the comprehensive evaluation matrix of the weld quality obtained above, the enterprise constructed the following parameter update equation:

[0232] ;

[0233] Using this equation, the enterprise continuously optimizes the detection parameters to ensure that all types of detection means are in the best state, thus obtaining more accurate and reliable weld quality evaluation results. After three rounds of iteration, the enterprise finally determined the following optimal parameter combination:

[0234] ;

[0235] With this optimized detection parameter setting, the enterprise can ensure the accuracy of the evaluation results, laying a foundation for subsequent real-time monitoring and early warning.

[0236] Finally, the enterprise established a real-time monitoring and abnormal early warning mechanism for the weld state according to step S60 of the present invention.

[0237] First, the enterprise defined three indicators reflecting the overall quality of the weld:

[0238] Integrity index , used to evaluate the internal defect situation of the weld;

[0239] Stress concentration index , used to evaluate the extreme stress state borne by the weld;

[0240] Fatigue damage index , used to evaluate the degree of fatigue damage of the weld.

[0241] For these three indicators, the enterprise respectively set safety thresholds:

[0242] ;

[0243] That is to say, when any one of the indicators exceeds the above threshold, the early warning mechanism will be triggered.

[0244] The enterprise collects and analyzes the above three indicator data in real time to form a weld state vector . Once it is found that the situation occurs, the system will immediately send out an early warning signal to remind the staff to check and repair in time. This can not only timely detect the abnormal weld quality, but also provide strong support for subsequent maintenance, ensuring the safe operation of the entire storage tank system.

[0245] It should be noted that the variable explanations involved in the present invention are shown in Table 1 below:

[0246] Table 1 Variable Explanation Table

[0247]

[0248] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for detecting weld seams of oil tank pontoon plate welding, characterized in that: The following steps are involved: S10, constructing a floating board motion characteristic matrix, including: collecting a height matrix of the floating board, recording real-time height data of measuring points at different positions; establishing a floating board attitude matrix, including pitch angle, roll angle, yaw angle and velocity components in three directions; obtaining a relative displacement matrix of the circumferential weld and the radial weld during the floating process of the floating board; S20, collecting the basic detection matrix of the weld, including: establishing an ultrasonic flaw detection feature matrix to record the ultrasonic reflection coefficient at different positions and different detection angles; constructing an X-ray imaging feature matrix to obtain the X-ray penetration at different positions and different exposure parameters; forming a weld surface morphology feature matrix, including weld contour and surface defect distribution information; S30, perform dynamic stress-strain analysis: establish a stress matrix including normal stress and shear stress components; construct a strain matrix including normal strain and shear strain components; calculate the stress field distribution considering the floating effect through the stress-strain correlation equation group; S40. Construct a comprehensive evaluation matrix for weld quality: perform weighted combination of ultrasonic flaw detection feature matrix, X-ray imaging feature matrix and stress-strain matrix; establish a weight coefficient matrix to ensure a reasonable proportion of various indicators; and form an evaluation matrix that reflects the overall quality status of the weld; S50, realizing adaptive adjustment of detection parameters: establishing a detection parameter matrix including ultrasonic angle, X-ray parameters and strain acquisition parameters; constructing a parameter update equation based on the weld quality evaluation matrix; realizing dynamic optimization and update of detection parameters; S60. Execute real-time monitoring and early warning: construct a weld state vector, including integrity indicators, stress concentration indicators and fatigue damage indicators; set a safety threshold matrix and establish early warning trigger conditions; realize real-time monitoring of weld state and abnormal early warning; The stress-strain correlation equation group includes a principal stress calculation equation, a shear stress calculation equation, a strain distribution equation, and a fatigue damage accumulation equation; The principal stress calculation equation is used to calculate the magnitude and direction of the principal stress at each point of the weld. The input includes the weld position coordinates, the height data of the floating ship plate, the attitude angle data and the velocity component. The output is the magnitude and direction of the principal stress at each measuring point. The shear stress calculation equation is used to determine the shear stress distribution of the weld, the input includes principal stress data, weld relative displacement data and material shear modulus, and the output is the shear stress component of each point of the weld and the maximum shear stress position; The strain distribution equation is used to calculate the deformation state of the weld, the input includes principal stress data, shear stress data, material elastic modulus and Poisson's ratio, and the output is the normal strain and shear strain distribution of the weld area; The fatigue damage accumulation equation is used to evaluate the fatigue state of the weld, and the input includes the number of stress cycles, stress amplitude, average stress and material fatigue parameters, and the output is the fatigue damage degree of each point of the weld; The parameter update equation is used to optimize the detection parameter setting, the input includes the current detection parameters, weld quality evaluation index, detection accuracy requirements and equipment performance constraints, and the output is the updated ultrasonic incident angle, X-ray exposure parameters and strain acquisition frequency; Among them, the integrity index is used to evaluate the internal defects of the weld; the stress concentration index is used to evaluate the extreme stress state that the weld is subjected to.

2. A method for detecting weld seams of oil tank pontoon plates according to claim 1, characterized in that: The buoyancy plate motion characteristic matrix is ​​expressed as follows: ; In the formula, Indicates The measurement point is The altitude value at the moment, in meters, is the number of measurement points, is the number of sampling moments; The posture matrix is ​​expressed as follows: ; In the formula, is the pitch angle, and the range is ; is the roll angle, and the range is ; is the yaw angle, and the range is ; are the velocity components in three directions, in meters per second; The relative displacement matrix is ​​expressed as follows: ; In the formula, is the relative displacement vector of the circumferential weld, is the radial weld relative displacement vector.

3. A method for detecting weld seams of oil tank pontoon plate welding according to claim 2, characterized in that: The principal stress calculation equation is expressed as follows: ; In the formula, is the stress in the x direction, is the stress in the y direction, is the shear stress, is the oil density, is the acceleration due to gravity, is the liquid level height, is the attitude influence coefficient, is the dynamic viscosity, where represents the principal stress, and the subscripts 1 and 2 represent the two principal stress values. is the maximum principal stress, is the minimum principal stress.

4. A method for detecting weld seams of oil tank pontoon plate welding according to claim 3, characterized in that: The shear stress calculation equation is expressed as follows: ; In the formula, is the shear modulus, is the shear strain, is the viscosity coefficient, is the coefficient of the second-order derivative term.

5. A method for detecting weld seams of oil tank pontoon plate welding according to claim 4, characterized in that: The strain distribution equation is expressed as follows: ; In the formula, is the elastic modulus, is Poisson's ratio, is the Kronecker symbol, are the components of the stress tensor, , is a subscript indicating the stress direction, where represents the normal direction of the stress surface, and the subscript j represents the direction of the stress; represents the trace of the stress tensor.

6. A method for detecting weld seams of oil tank pontoon plate welding according to claim 5, characterized in that: The fatigue damage accumulation equation is expressed as follows: ; In the formula, is the actual number of cycles, is the fatigue life, is the injury index, is the correction factor, is the stress amplitude, is the fatigue limit, is the material constant.

7. A method for detecting weld seams of oil tank pontoon plate welding according to claim 6, characterized in that: The parameter update equation is expressed as follows: ; In the formula, is the current parameter vector, is the learning rate, is the performance indicator function, is the second-order derivative weight, is a random disturbance term.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed in a computer, they are used to execute the weld detection method for welding of a petroleum storage tank pontoon plate according to any one of claims 1 to 7.

9. A weld detection system for welding of oil tank pontoon plates, characterized in that: Contains the computer-readable storage medium of claim 8.

Images