A method and system for optimizing stress concentration data of high-pressure petrochemical containers
By using adaptive mesh division and decomposition amount evaluation technology in the three-dimensional model of petrochemical pressure vessels, the stress risk area is identified and optimized, and the problem of insufficient accuracy in the stress concentration area in the traditional method is solved, and the safety and efficiency of the design are improved.
Patent Information
- Application Number
- CN202410986025.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-23
AI Technical Summary
Traditional design and optimization methods are difficult to accurately calculate the insufficient accuracy of petrochemical pressure vessels in stress-concentrated areas, resulting in safety hazards and material fatigue. The finite element analysis is computationally large and has low efficiency, making it difficult to quickly respond to design and optimization needs.
By establishing a three-dimensional model, adaptive mesh division technology is used to dynamically adjust the grid density, calculate stress and strain tensors, use positive and inverse decomposition amounts to evaluate the load-voltage combined risk amount, identify and optimize stress risk areas, and perform rounding angles Edge treatment to reduce stress concentration.
It improves the accuracy and reliability of stress analysis, reduces the risk of material fatigue and fracture, and improves the safety and efficiency of the design.
Smart Images

Figure CN118673724B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of data processing and data optimization, and in particular relates to a method and system for optimizing stress concentration resistance data of a high-pressure petrochemical container. Background Art
[0002] The three-dimensional model design of petrochemical pressure vessels needs to meet the stress requirements for high pressure resistance. In the three-dimensional model, it is generally believed that stress is the load per unit area, and the relationship between stress and load can be expressed as a positive correlation between stress and load and a negative correlation between stress and the area under stress. However, after the introduction of the normal stress or normal stress perpendicular to the cross section, it was found that the external force on some materials does not change with time during operation. At this time, the internal stress remains unchanged, which is called static stress. There are also some materials whose external forces change periodically with time. At this time, the internal stress also changes periodically with time, which is called alternating stress. In addition, in the process of stress increasing with the increase of external force, the material will cause local increase in stress due to the change of cross-sectional size. For brittle materials with uniform structure, stress concentration will greatly reduce the strength of the component, which needs to be determined by mechanical testing of the material. Among them, based on the material properties, the load caused by the force on the three-dimensional model may produce very small deformation. When performing finite element analysis, it is necessary to calculate the stress and strain tensors in each load step to evaluate the performance of the structure under different loads. This first requires the establishment of a finite element model of the structure, dividing the structure into multiple small units, and then performing mechanical modeling on each small unit. In each load step, the force conditions of each small unit need to be calculated to determine its stress and strain tensors. Before performing finite element analysis, the loads and boundary conditions of the structure need to be determined. The load can be an externally applied force or pressure, and the boundary conditions can be the support or constraints of the structure. According to the geometry and force conditions of each small unit, the strain tensor of each small unit can be calculated by mathematical formulas. The strain tensor describes the deformation of the material under force. According to the material properties and strain tensor of each small unit, the stress tensor of each small unit can be calculated by the mechanical properties of the material itself, and the stress tensor describes the stress of the material under force. Strain is related to the relative displacement of particles inside the object. This relative displacement is called deformation, which does not include the position change of the object due to translational and rotational rigid body motion. Different mathematical models can be used to distinguish between small and large strain conditions, namely infinitesimal strain theory and finite / large strain theory. In practice, there are many ways to calculate minimum strain, the commonly used ones are Cauthy strain and Green strain.During the design process, the three-dimensional contour of petrochemical pressure vessels is generally designed to be relatively rounded, such as a cylinder. However, the bottom connecting edge of the cylinder or the assembly connecting area of the container components still inevitably have relatively sharp edges. These subtle details often accumulate stress concentration risks unmonitored, especially in the actual industrial container design templates and those practical containers with low production costs. This situation is difficult to avoid and often brings safety hazards. Optimization on the three-dimensional model can solve the problem fundamentally from the template, facilitating subsequent large-scale production applications.
[0003] Petrochemical pressure vessels are widely used in industrial production and need to withstand high pressure and complex stress conditions. In traditional design and optimization methods, the empirical design method is mainly based on the experience of engineers and industry standards. This method relies on the experience of engineers, is difficult to quantify and standardize, and has great uncertainty and safety hazards. The analytical method uses classical mechanics theory to calculate the stress distribution of the container through analytical solutions, but the analytical method is only applicable to containers with simple geometric shapes and cannot cope with complex three-dimensional structures. Finite element analysis (FEA) uses the finite element method to numerically simulate the container and calculate the stress and strain distribution. However, the calculation accuracy of traditional finite element analysis in stress concentration areas is insufficient, and the calculation amount is large and the efficiency is low. Due to the inadequate optimization of mesh division, the calculation results may be inaccurate. The material testing method guides the design by determining the performance of the material under different stress conditions through mechanical tests. However, the test cost is high, the cycle is long, and the test results have a certain degree of discreteness, which makes it difficult to fully cover all use conditions. For example, the traditional method of prestressed steel pipes described in the patent document with publication number CN101498384B is not effective in identifying and processing stress concentration areas, which easily leads to fatigue and fracture of the container in high stress concentration areas. A wall shear stress optimization method described in a patent document with publication number CN109919916B requires a lot of calculation and test time, and it is difficult to quickly respond to design and optimization needs. It is difficult to consider stress distribution, material cost and manufacturing process at the same time, resulting in insufficient comprehensive performance of the design scheme. Due to insufficient optimization of mesh division, the calculation accuracy of traditional finite element analysis methods in stress concentration areas is insufficient, affecting the reliability of the design. Summary of the invention
[0004] The purpose of the present invention is to propose a method and system for optimizing stress concentration data of a high-pressure petrochemical container to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.
[0005] The present invention provides a method and system for optimizing stress concentration data of high-pressure petrochemical containers, which can establish a three-dimensional model and perform grid division; determine loads and boundary conditions; calculate stress and strain tensors; dynamically adjust grid density using adaptive grid division technology; calculate stress load vectors of edge points through positive and negative order decomposition, and evaluate the combined load and pressure risk; identify and optimize edges with stress risks according to the risk difference of edge points in each edge. Finally, rounded edges are processed for high-risk edges.
[0006] In order to achieve the above object, according to one aspect of the present invention, a method for optimizing stress concentration data of a high-pressure petrochemical container is provided, the method comprising the following steps:
[0007] Obtaining a three-dimensional model of a petrochemical pressure vessel, marking edges on the three-dimensional model and marking coordinate points on each edge as edge marking points;
[0008] Setting a plurality of different load steps for the three-dimensional model, and obtaining stress values of edge points on each edge corresponding to each load step;
[0009] The multiple different load steps are a number of load values that start from a non-zero load and then gradually increase multiple times;
[0010] The stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point, and the stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount;
[0011] The positive order load-pressure decomposition and the reverse order load-pressure decomposition of the edge points on each edge are condensed to obtain the load-pressure combined risk amount;
[0012] In each edge, each edge point is compared according to the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point, and the load-pressure combined with the risk amount of each edge point in the edge are compared again to obtain the edges with stress risks in the three-dimensional model, and the edges with stress risks in the three-dimensional model are rounded.
[0013] Furthermore, in the plurality of different load steps, the starting load is non-zero, and the starting load corresponding to each mark point remains consistent, and the increment of the load step corresponding to each mark point also remains consistent.
[0014] Furthermore, the three-dimensional model of the petrochemical pressure vessel is composed of a plurality of different coordinate points, and the three-dimensional model can be used to represent the three-dimensional geometric features of points, lines and surfaces of the petrochemical pressure vessel.
[0015] Furthermore, in each edge, each edge point is compared according to the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point, and the load-pressure combined risk amount of each edge point in the edge is compared again to obtain the edges with stress risk in the three-dimensional model, which are specifically:
[0016] In each edge, the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of each edge point are combined to obtain the double order load-pressure combination amount of each edge point;
[0017] And in each edge, calculate the average value of the cosine similarity between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point similarity of the edge point, and calculate the average value of the Euclidean distance between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point distance of the edge point;
[0018] In each edge, the local comparative risk value of each edge point is obtained by combining the edge point similarity and edge point distance values of each edge point, wherein the local comparative risk value is a value that can be used to measure the risk sensitivity of each edge point in an edge compared with the other edge points;
[0019] In each edge, the difference between the combined load-pressure risk and the local comparative risk of each edge point is calculated as the risk difference of the edge point;
[0020] According to the risk amount difference of the edge points in each edge, the edges with stress risk in the three-dimensional model are obtained.
[0021] In each edge, the positive and negative load-pressure decompositions of each edge point are compared, and the load-pressure combined with the risk amount is combined to identify the edges with stress risks in the three-dimensional model. Quantifying the stress distribution of edge points in different load directions by positive and negative decompositions helps to identify stress concentration areas, and is used to quantify the similarity and distance between edge points, which can provide quantitative analysis of local stress distribution. Combining similarity and distance to obtain the risk amount of edge points can better quantitatively measure the stress sensitivity of each point, which is conducive to accurately identifying edges with greater stress risks in the three-dimensional model, improving the accuracy and reliability of stress analysis, and reducing structural failures caused by stress concentration.
[0022] Furthermore, the stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point, and the stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount, which are specifically:
[0023] For each boundary point, the array of stress values corresponding to each load step of the boundary point is used as the stress load vector of the boundary point, wherein the stress load vector of the boundary point is decomposed along the positive order of the increasing direction of the load step to obtain the positive order load-pressure decomposition, which can be specifically:
[0024] Traverse the stress load vector, if the dimension before the current dimension is not empty, then for the stress value of the current dimension, calculate the maximum value of the stress values of the dimensions before the current dimension as the positive sequence stress starting peak value of the current dimension, and use the ratio of the stress value of the current dimension to the positive sequence stress starting peak value of the current dimension as the positive sequence load-pressure decomposition value of the current dimension, thereby calculating the positive sequence load-pressure decomposition values of each dimension except the first dimension in the stress load vector, and use the array composed of the positive sequence load-pressure decomposition values of each dimension except the first dimension as the positive sequence load-pressure decomposition amount;
[0025] Among them, the stress load vector of each edge point is decomposed in the reverse order of the increasing direction of the load step to obtain the reverse load and pressure decomposition, specifically:
[0026] When traversing the stress load vector, if the dimension after the current dimension is not empty, for the stress value of the current dimension, calculate the maximum value of the stress values of each dimension after the current dimension as the starting peak value of the subsequent stress of the current dimension, and use the ratio of the stress value of the current dimension to the starting peak value of the reverse stress of the current dimension as the reverse load-pressure decomposition value of the current dimension, thereby calculating the reverse load-pressure decomposition values of each dimension in the stress load vector except the first dimension, and use the array composed of the reverse load-pressure decomposition values of each dimension except the first dimension as the reverse load-pressure decomposition amount.
[0027] The stress load vector of the boundary point is calculated, and the load-pressure decomposition is obtained by decomposing it in the positive and negative order. The stress load vector represents the stress distribution of the boundary point under different load steps, and the positive and negative order decompositions are decomposed in different directions of the load step to obtain the positive and negative order load-pressure decompositions, providing a dual perspective analysis of the stress change trend. A two-way decomposition method is provided to accurately quantify the stress of the boundary point in different load directions. The accuracy and comprehensiveness of stress decomposition are improved through two-way analysis, providing reliable data for subsequent risk analysis.
[0028] Furthermore, the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of the edge points on each edge are condensed to obtain the load-pressure combined risk amount, which is specifically:
[0029] For the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point on each edge, respectively, the cosine similarity between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition similarity of the edge point, and the Euclidean distance between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition distance of the edge point;
[0030] The load-pressure combined risk of the boundary point is obtained by combining the similarity of the forward and reverse order load-pressure decomposition quantities and the distance of the forward and reverse order load-pressure decomposition quantities of the boundary point.
[0031] The similarity and distance between the positive and reverse load-pressure decompositions of each edge point are calculated, and these values are combined to obtain the combined load-pressure risk, which helps to quantify the relationship between the positive and reverse load-pressure decompositions and reveal the inherent pattern of stress change. Among them, the combined load-pressure risk combines similarity and distance to comprehensively evaluate the stress risk of edge points and provide a comprehensive risk assessment index. It can help quantitatively measure the stress risk of edge points and enhance the accuracy and comprehensiveness of risk assessment.
[0032] Furthermore, according to the risk difference of the edge points in each edge, the edge with stress risk in the three-dimensional model is obtained, specifically:
[0033] According to the risk amount difference of the edge marking points in each edge, rounded edge processing is performed on each edge respectively, wherein, if the risk amount difference of the edge marking points in an edge is greater than a threshold value, it is judged that the stress risk of the edge under pressure conditions is greater, and rounded edge processing is performed on the edge with the greater stress risk in the three-dimensional model, and the threshold value is a value obtained by statistically analyzing the risk amount difference of each edge marking point.
[0034] According to the risk difference of the edge points in each edge, the edges with stress risk are rounded. The risk difference can quantify the risk level of each edge point and is used to identify high-risk areas. The rounded edge treatment reduces stress concentration and optimizes the structural design by rounding the high-risk edges. In this way, the risk of stress concentration areas is reduced through structural optimization. This improves the structure's ability to resist stress concentration, reduces the risk of material fatigue and fracture, and improves overall safety and reliability.
[0035] The present invention also provides a stress concentration data optimization system for a high-pressure petrochemical container, the stress concentration data optimization system for a high-pressure petrochemical container comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implementing the steps in the stress concentration data optimization method for a high-pressure petrochemical container when executing the computer program, and controlling the steps accordingly, the stress concentration data optimization system for a high-pressure petrochemical container can be run in computing devices such as desktop computers, laptop computers, mobile phones, PDAs, and cloud data centers, and the executable systems may include, but are not limited to, processors, memories, and server clusters, and the processor executing the computer program runs in the following system units:
[0036] A three-dimensional marking unit, used to obtain a three-dimensional model of a petrochemical pressure vessel, mark the edges on the three-dimensional model and mark the coordinate points on each edge as edge marking points;
[0037] A stress load unit, used to set a plurality of different load steps for the three-dimensional model, and obtain the stress value of each edge point corresponding to each load step;
[0038] A data decomposition unit is used to form a stress load vector of each boundary point with the stress value of each boundary point corresponding to each load step, and to decompose the stress load vector of each boundary point along the positive sequence and reverse sequence of the increasing direction of the load step to obtain a positive sequence load-pressure decomposition amount and a reverse sequence load-pressure decomposition amount;
[0039] A data condensation unit is used to condense the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of the edge punctuation points on each edge to obtain the load-pressure combined risk amount;
[0040] The optimization processing unit is used to compare each edge point in each edge according to the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of each edge point, and compare again in combination with the load-pressure combined with the risk amount of each edge point in the edge, to obtain the edges with stress risks in the three-dimensional model, and perform rounded edge processing on the edges with stress risks in the three-dimensional model.
[0041] The beneficial effects of the present invention are as follows: the present invention provides a method and system for optimizing stress concentration data of a high-pressure petrochemical container, wherein the stress value of each edge point corresponding to each load step forms the stress load vector of the edge point, and the stress load vector of each edge point is decomposed along the load step direction to obtain the positive load-pressure decomposition and the reverse load-pressure decomposition; the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point on each edge are condensed to obtain the load-pressure combined risk; within each edge, the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point are compared, and the edges with stress risks in the three-dimensional model are identified in combination with the load-pressure combined risk, and the edges with stress risks are rounded. This is beneficial to reducing the probability of material fatigue and fracture risks caused by stress concentration in the three-dimensional model design of petrochemical pressure vessels, and helps to quickly and easily identify edges with stress concentration risks in the three-dimensional model. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:
[0043] Figure 1 Shown is a flow chart of a method for optimizing stress concentration data of a high-pressure petrochemical container;
[0044] Figure 2 Shown is a system structure diagram of a stress concentration resistance data optimization system for a high-pressure petrochemical container. DETAILED DESCRIPTION
[0045] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.
[0046] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.
[0047] like Figure 1The flowchart of the stress concentration data optimization method for a high pressure petrochemical container according to the present invention is shown below. Figure 1 A method and system for optimizing stress concentration data of a high-pressure petrochemical container according to an embodiment of the present invention will be described.
[0048] The present invention proposes a method for optimizing stress concentration data of a high-pressure petrochemical container, the method specifically comprising the following steps:
[0049] Obtaining a three-dimensional model of a petrochemical pressure vessel, marking edges on the three-dimensional model and marking coordinate points on each edge as edge marking points;
[0050] Setting a plurality of different load steps for the three-dimensional model, and obtaining stress values of edge points on each edge corresponding to each load step;
[0051] The stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point, and the stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount;
[0052] The positive order load-pressure decomposition and the reverse order load-pressure decomposition of the edge points on each edge are condensed to obtain the load-pressure combined risk amount;
[0053] In each edge, each edge point is compared according to the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point, and the load-pressure combined with the risk amount of each edge point in the edge are compared again to obtain the edges with stress risks in the three-dimensional model, and the edges with stress risks in the three-dimensional model are rounded.
[0054] The multiple different load steps are a number of load values that start from a non-zero load and then gradually increase multiple times.
[0055] In some embodiments, a three-dimensional model can be established and meshed; loads and boundary conditions can be determined; stress and strain tensors can be calculated through finite element analysis, etc.; mesh density can be dynamically adjusted using adaptive meshing technology; stress load vectors of edge points can be calculated through positive and negative order decomposition, and the combined load and pressure risk can be evaluated; edges with stress risks can be identified and optimized based on the risk difference of edge points in each edge. Finally, high-risk edges can be rounded. In the process of meshing, the three-dimensional model can also be divided into multiple different sub-areas according to boundary conditions, etc., and the edges across each sub-area can also be divided in the division of sub-areas. Each sub-area should also subdivide the edges of the three-dimensional model into several small edges.
[0056] Furthermore, in the plurality of different load steps, the value of the starting load in the increment is greater than zero, and the starting load corresponding to each mark point remains consistent, and the increment of the load steps corresponding to each mark point is also consistent.
[0057] Furthermore, the three-dimensional model of the petrochemical pressure vessel is composed of a plurality of different coordinate points, and the three-dimensional model can be used to represent the three-dimensional geometric features of points, lines and surfaces of the petrochemical pressure vessel.
[0058] In some embodiments, lines that can be used to represent the contour features of the three-dimensional model, angular edge lines, etc. can be used as edges on the three-dimensional model. Sampling is performed on each edge of the three-dimensional model to obtain multiple points as edge marking points on the edge. The number of edge marking points on an edge should be a countable, finite positive integer.
[0059] Furthermore, the stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point, and the stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount, which are specifically:
[0060] In some embodiments, for each boundary point, an array consisting of stress values corresponding to each load step of the boundary point is used as the stress load vector of the boundary point;
[0061] For a boundary point on the three-dimensional model of a petrochemical pressure vessel, there are k load steps corresponding to different external forces. The serial number of each load step is denoted as d, and d belongs to the interval 1 to k, where the load step with serial number d is step(d), the load step with serial number 1 is step(1), and the load step with serial number k is step(k). The stress value corresponding to the load step step(d) is strs(d), the stress value with serial number 1 is strs(1), and the stress value with serial number k is strs(k). For example, when there are 5 load steps, namely F=10N, 20N, 40N, 60N and 80N, the specific value of k is 5, and the specific range of d is 1-5.
[0062] Among them, the stress load vector of the edge point is decomposed along the positive order of the increasing direction of the load step to obtain the positive order load and pressure decomposition, which can be specifically:
[0063] In some embodiments, the stress load is respectively along the increasing direction of the load step, that is, along the increasing direction from small to large load steps from step (1) to step (k), for example, from 10N to 80N, the number of dimensions of the positive sequence load-pressure decomposition is reduced by 1 compared with the number of dimensions of the stress load vector, that is, the number of dimensions of the positive sequence load-pressure decomposition is k-1, and consistent with the dimension of the stress load vector, the traversal range of d in the positive sequence load-pressure decomposition is limited to the interval of 2 to k, and the maximum value of the stress value of the stress load vector in the dimensions before the current dimension with the sequence number d is calculated as its corresponding sequence stress starting peak value strs(d)aug. In order to highlight the ratio of the stress value strs(d) of the current dimension to the positive sequence stress starting peak value strs(d)aug of the current dimension, strs(d) and strs(d)aug can also be processed by an activation function or the like;
[0064] The positive sequence stress starting peak value indicates the stress peak value starting from the first dimension to the current dimension in the positive sequence along the increasing direction of the load step. The ratio of the positive sequence stress starting peak value to each dimension can be used to compare the subsequent stress fluctuations corresponding to the positive linear correlation distribution relationship in the positive sequence process of the load step increasing with the external force, so that the positive sequence load-pressure decomposition is conducive to extracting the sensitivity of each edge point on the three-dimensional model of the high-pressure vessel to the load increase, and helping to find out the places with sensitive risks to the load in the model design for optimization;
[0065] Traversing the stress load vector, if the dimension before the current dimension is not empty, for the stress value of the current dimension, calculate the maximum value of the stress values of each dimension before the current dimension as the positive-sequence stress starting peak value of the current dimension, and use the ratio of the stress value of the current dimension to the positive-sequence stress starting peak value of the current dimension as the positive-sequence load-pressure decomposition value of the current dimension, thereby calculating the positive-sequence load-pressure decomposition values of each dimension in the stress load vector except the first dimension, and use the array composed of the positive-sequence load-pressure decomposition values of each dimension except the first dimension as the positive-sequence load-pressure decomposition amount.
[0066] Among them, the stress load vector of each edge point is decomposed in the reverse order of the increasing direction of the load step to obtain the reverse load and pressure decomposition, which can be specifically:
[0067] In some embodiments, the number of dimensions of the reverse load-pressure decomposition is reduced by 1 compared to the number of dimensions of the stress load vector, that is, the number of dimensions of the reverse load-pressure decomposition is also k-1. In a similar manner, the traversal range of d in the reverse load-pressure decomposition is limited to the interval of 1 to k-1. The reverse load-pressure decomposition is composed of the reverse load-pressure decomposition values corresponding to the first dimension to the second last dimension in the stress load vector, wherein the first dimension to the second last dimension in the stress load vector correspond to the dimensions numbered 1 to k-1, and the reverse load-pressure decomposition value is calculated by comparing the stress value corresponding to the dimension whose last dimension in the stress load vector is not empty with the stress values of the subsequent dimensions;
[0068] When traversing the stress load vector, if the dimension after the current dimension is not empty, for the stress value of the current dimension, calculate the maximum value of the stress values of each dimension after the current dimension as the starting peak value of the subsequent stress of the current dimension, and use the ratio of the stress value of the current dimension to the starting peak value of the reverse stress of the current dimension as the reverse load-pressure decomposition value of the current dimension, thereby calculating the reverse load-pressure decomposition values of each dimension in the stress load vector except the first dimension, and use the array composed of the reverse load-pressure decomposition values of each dimension except the first dimension as the reverse load-pressure decomposition amount.
[0069] The reverse stress starting peak indicates the most prominent stress peak from the current dimension to the last dimension in the reverse order of the increasing direction of the load step.
[0070] The ratio of the initial peak value of the reverse stress to each dimension can be used to predictively measure the subsequent stress fluctuations in the time series, which corresponds to the reverse linear correlation distribution relationship in the reverse process of the load intensity decreasing from the last dimension to the current dimension in the load step, so that the reverse load-pressure decomposition is conducive to extracting the sensitivity of each edge point on the three-dimensional model of the high-pressure container to the load drop, and better help find out the places where there is a risk of sensitivity to the load in the model design for optimization from the aspect of load drop.
[0071] Furthermore, the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of the edge points on each edge are condensed to obtain the load-pressure combined risk amount, which is specifically:
[0072] For the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point on each edge, respectively, the cosine similarity between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition similarity of the edge point, and the Euclidean distance between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition distance of the edge point;
[0073] The load-pressure combined risk of the boundary point is obtained by combining the similarity of the forward and reverse order load-pressure decomposition quantities and the distance of the forward and reverse order load-pressure decomposition quantities of the boundary point.
[0074] Among them, the similarity of the positive and negative load and pressure decomposition can be used to represent the directional deviation between the positive and negative load and pressure decomposition of a boundary point. The deviation in this direction is used to measure the deviation in the direction of change in the linear relationship between the stress and load steps in the increasing and decreasing directions. The positive and negative load and pressure decomposition distance of a boundary point can be used to represent the extended distance between the positive and negative load and pressure decomposition of a boundary point. This extended distance is used to measure the mathematical expectation of the increasing and decreasing directions between the stress and load steps after vector serialization. The use of the deviation or mathematical expectation of the change direction in the quantization space in the increasing and decreasing directions between the stress and load steps alone will be biased, but the combination of the deviation and mathematical expectation of the change direction in the quantization space in the increasing and decreasing directions between the stress and load steps is conducive to accurately measuring the size and development direction of the safety risk of each boundary point for stress changes, and better identifying the boundary points with sensitive risks to loads in model design.
[0075] In some embodiments, preferably, the numerical values of the similarity of the positive and negative load-pressure decomposition amounts and the distance of the positive and negative load-pressure decomposition amounts of the edge point can be multiplied, and the product obtained by combining them can be used as the load-pressure combined risk amount of the edge point; less preferably, in previous development experiments, the average value of the similarity of the positive and negative load-pressure decomposition amounts and the distance of the positive and negative load-pressure decomposition amounts of the edge point was used as the load-pressure combined risk amount of the edge point, but in the experimental data statistics, it was found that although the less preferred implementation method can also realize the combination of the numerical values of the similarity of the positive and negative load-pressure decomposition amounts and the distance of the positive and negative load-pressure decomposition amounts of the edge point, it is very easy to be inaccurate when the vector dimension increases. For example, the number of dimensions of the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount will increase with the increase of load steps, which will lead to inaccurate numerical fluctuations.
[0076] Furthermore, in each edge, each edge point is compared according to the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point, and the load-pressure combined risk amount of each edge point in the edge is compared again to obtain the edges with stress risk in the three-dimensional model, which are specifically:
[0077] In each edge, the positive order load-pressure decomposition and the reverse order load-pressure decomposition of each edge point are combined to obtain the double order load-pressure combination of each edge point:
[0078] In some embodiments, it is less preferred that the vectors obtained by multiplying the positive order load-pressure decomposition and the reverse order load-pressure decomposition of each edge point by the dimension point can be used as the double order load-pressure combination of each edge point. This less preferred implementation should be used when the simplicity of data calculation is required.
[0079] In some embodiments, preferably, the numerical values of each dimension in the positive-sequence load-pressure decomposition and the reverse-sequence load-pressure decomposition of each edge point can be averaged by dimension and the array composed of the average values can be used as the double-sequence load-pressure combination of each edge point. Such a combination method after averaging by dimension is more conducive to balancing the change direction deviation and mathematical expectation in the quantization space in both increasing and decreasing directions between the positive-sequence load-pressure decomposition and the reverse-sequence load-pressure decomposition of each edge point. The combination method of performing dot multiplication of the positive-sequence load-pressure decomposition and the reverse-sequence load-pressure decomposition described in the less preferred embodiment is simple but easily loses the data characteristics of the directionality and ductility of the positive-sequence load-pressure decomposition and the reverse-sequence load-pressure decomposition.
[0080] And in each edge, calculate the average value of the cosine similarity between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point similarity of the edge point, and calculate the average value of the Euclidean distance between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point distance of the edge point;
[0081] In some embodiments, within each edge, for each edge punctuation point in this edge, the cosine similarities between the double-sequence load-pressure combination of the selected edge punctuation point and the double-sequence load-pressure combination of the remaining edge punctuation points in this edge can be calculated, and then the arithmetic mean of the numerical values of each cosine similarity can be calculated, and the arithmetic mean of each cosine similarity can be used as the corresponding edge punctuation point similarity of the selected edge punctuation point in this edge; and the Euclidean distances between the double-sequence load-pressure combination of the selected edge punctuation point and the double-sequence load-pressure combination of the remaining edge punctuation points in this edge can be calculated, and then the arithmetic mean of the numerical values of each Euclidean distance can be calculated, and the arithmetic mean of each Euclidean distance can be used as the corresponding edge punctuation point distance of the selected edge punctuation point in this edge.
[0082] In each edge, the local comparative risk value of each edge point is obtained by combining the edge point similarity and edge point distance values of each edge point, wherein the local comparative risk value is a value that can be used to measure the risk sensitivity of each edge point in an edge compared with the other edge points:
[0083] In some embodiments, corresponding to the calculation method of the load-pressure combined risk amount of the boundary marker point, the local comparative risk amount can also be implemented as follows: preferably, the values of the boundary marker point similarity and the boundary marker point distance of each boundary marker point can be averaged, and the obtained average value is used as the local comparative risk amount of the boundary marker point; less preferably, the values of the boundary marker point similarity and the boundary marker point distance of each boundary marker point can be multiplied, and the product is used as the local comparative risk amount of the boundary marker point.
[0084] In each edge, the difference between the combined load-pressure risk and the local comparative risk of each edge point is calculated as the risk difference of the edge point;
[0085] In some embodiments, for each edge point within an edge, the absolute value of the numerical difference between the load-pressure combined risk amount of each edge point and the local contrast risk amount of the edge point can be obtained as the risk amount difference of the edge point.
[0086] According to the risk amount difference of the edge points in each edge, the edges with stress risk in the three-dimensional model are obtained.
[0087] Furthermore, according to the risk difference of the edge points in each edge, the edge with stress risk in the three-dimensional model is obtained, specifically:
[0088] According to the risk amount difference of the edge marking points in each edge, rounded edge processing is performed on each edge respectively, wherein, if the risk amount difference of the edge marking points in an edge is greater than a threshold value, it is judged that the stress risk of the edge under pressure conditions is greater, and rounded edge processing is performed on the edge with the greater stress risk in the three-dimensional model, and the threshold value is a value obtained by statistically analyzing the risk amount difference of each edge marking point.
[0089] In some embodiments, the risk amount difference of the edge punctuation points in each edge is counted, for example, including but not limited to counting the average, cumulative value, integral result and / or mathematical expectation of the risk amount difference of the edge punctuation points in each edge as the statistical value after the risk amount difference of the edge punctuation points in each edge is counted;
[0090] Preferably, the edges are sorted according to the statistical values of the edges. The edges with larger statistical values are more likely to break when facing the stress caused by the high-pressure environment. The edges with original edges can be rounded to make the edges transition into rounded edges. Because the edges with edges are prone to stress concentration, the edges with edges can be converted into rounded edges with smooth transitions using tools including but not limited to PythonOCC, OpenCASCADE and / or CAD, etc., so as to enhance the durability and fatigue resistance of the container and reduce the occurrence of cracks on the container.
[0091] Preferably, according to the statistical values in each edge, the arithmetic mean of the statistical values in each edge is used as a threshold, and the edges whose statistical values are greater than or equal to the threshold can be used as edges to be optimized, and rounding processing (rounded edge processing) is performed on the edges to be optimized;
[0092] Alternatively, it is also preferred that the edge points in each edge are compared with the threshold, the edge points in each edge that are greater than or equal to the threshold are used as edge points to be optimized, and the shortest line segment in each edge that can contain all the edge points in the edge is rounded. The rounding can be performed by inputting a rounding radius or an expression, and the rounding method includes a smooth transition and / or a linear transition. Then, the three-dimensional model of the petrochemical pressure vessel after the rounded edge processing is saved and output;
[0093] Preferably, when there are no less than two points to be optimized on an edge, the portion of the angular edge between the two points to be optimized whose values exceed the maximum threshold, which is ranked first and second, is used as the edge to be optimized, and then the edge to be optimized is rounded.
[0094] The stress concentration resistance data optimization system for high-pressure petrochemical containers runs on any computing device such as a desktop computer, a laptop computer, a mobile phone, a PDA or a cloud data center. The computing device includes: a processor, a memory and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps in the stress concentration resistance data optimization method for high-pressure petrochemical containers are implemented. The executable system may include, but is not limited to, a processor, a memory, and a server cluster.
[0095] The embodiment of the present invention provides a stress concentration resistance data optimization system for a high-pressure petrochemical container, such as Figure 2 As shown, a stress concentration resistance data optimization system for a high-pressure petrochemical container of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned stress concentration resistance data optimization method for a high-pressure petrochemical container are implemented. The processor executes the computer program to run in the following system units:
[0096] A three-dimensional marking unit, used to obtain a three-dimensional model of a petrochemical pressure vessel, mark the edges on the three-dimensional model and mark the coordinate points on each edge as edge marking points;
[0097] A stress load unit, used to set a plurality of different load steps for the three-dimensional model, and obtain the stress value of each edge point corresponding to each load step;
[0098] A data decomposition unit is used to form a stress load vector of each boundary point with the stress value of each boundary point corresponding to each load step, and to decompose the stress load vector of each boundary point along the positive sequence and reverse sequence of the increasing direction of the load step to obtain a positive sequence load-pressure decomposition amount and a reverse sequence load-pressure decomposition amount;
[0099] A data condensation unit is used to condense the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of the edge punctuation points on each edge to obtain the load-pressure combined risk amount;
[0100] The optimization processing unit is used to compare each edge point in each edge according to the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of each edge point, and compare again in combination with the load-pressure combined with the risk amount of each edge point in the edge, to obtain the edges with stress risks in the three-dimensional model, and perform rounded edge processing on the edges with stress risks in the three-dimensional model.
[0101] Among them, preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.
[0102] Among them, dimensionless numerical calculations are used between physical quantities of different units.
[0103] The stress concentration data optimization system for high-pressure petrochemical containers can be run on computing devices such as desktop computers, laptops, mobile phones, PDAs, and cloud data centers. The stress concentration data optimization system for high-pressure petrochemical containers includes, but is not limited to, processors and memories. Those skilled in the art can understand that the example is only an example of a stress concentration data optimization method and system for high-pressure petrochemical containers, and does not constitute a limitation on a stress concentration data optimization method and system for high-pressure petrochemical containers. It can include more or fewer components than the example, or a combination of certain components, or different components. For example, the stress concentration data optimization system for high-pressure petrochemical containers can also include input and output devices, network access devices, buses, etc.
[0104] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete component gate circuits or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the stress-resistant concentrated data optimization system for high-pressure petrochemical containers, and uses various interfaces and lines to connect the various sub-regions of the entire stress-resistant concentrated data optimization system for high-pressure petrochemical containers.
[0105] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the stress concentration data optimization method and system for a high-pressure petrochemical container by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0106] The present invention provides a method and system for optimizing stress concentration data of high-pressure petrochemical containers. The stress value of each edge point corresponding to each load step forms the stress load vector of the edge point. The stress load vector of each edge point is decomposed along the load step direction to obtain the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount; the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of the edge point on each edge are condensed to obtain the load-pressure combined risk amount; within each edge, the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of each edge point are compared, and the edges with stress risks in the three-dimensional model are identified in combination with the load-pressure combined risk amount, and the edges with stress risks are rounded. This is beneficial to reducing the probability of material fatigue and fracture risks caused by stress concentration in the three-dimensional model design of petrochemical pressure vessels, and helps to quickly and easily identify edges with stress concentration risks in the three-dimensional model.
[0107] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.
Claims
1. A method for optimizing stress concentration data of a high-pressure petrochemical container, characterized in that: The method comprises the following steps: Obtaining a three-dimensional model of a petrochemical pressure vessel, marking edges on the three-dimensional model and marking coordinate points on each edge as edge marking points; Setting a plurality of different load steps for the three-dimensional model, and obtaining stress values of edge points on each edge corresponding to each load step; The multiple different load steps are a number of load values that start from a non-zero load and then gradually increase multiple times; The stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point, and the stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount; The positive order load-pressure decomposition and the reverse order load-pressure decomposition of the edge points on each edge are condensed to obtain the load-pressure combined risk amount; In each edge, each edge point is compared according to the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of each edge point, and the load-pressure combined with the risk amount of each edge point in the edge are compared again to obtain the edge with stress risk in the three-dimensional model, and the edge with stress risk in the three-dimensional model is rounded; Among them, in each edge, each edge point is compared according to the positive load-pressure decomposition amount and the reverse load-pressure decomposition amount of each edge point, and the load-pressure combined risk amount of each edge point in the edge is compared again to obtain the edges with stress risk in the three-dimensional model, which are specifically: In each edge, the positive order load-pressure decomposition amount and the reverse order load-pressure decomposition amount of each edge point are combined to obtain the double order load-pressure combination amount of each edge point; And in each edge, calculate the average value of the cosine similarity between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point similarity of the edge point, and calculate the average value of the Euclidean distance between the double-order load-pressure combination of each edge point and the double-order load-pressure combination of the other edge points as the edge point distance of the edge point; In each edge, the local comparative risk value of each edge point is obtained by combining the edge point similarity and edge point distance values of each edge point, wherein the local comparative risk value is a value that can be used to measure the risk sensitivity of each edge point in an edge compared with the other edge points; In each edge, the difference between the combined load-pressure risk and the local comparative risk of each edge point is calculated as the risk difference of the edge point; According to the risk amount difference of the edge points in each edge, the edge with stress risk in the three-dimensional model is obtained; Among them, the positive order load-pressure decomposition and the reverse order load-pressure decomposition of the edge points on each edge are condensed to obtain the load-pressure combined risk, which is: For the positive load-pressure decomposition and the reverse load-pressure decomposition of each edge point on each edge, respectively, the cosine similarity between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition similarity of the edge point, and the Euclidean distance between the positive load-pressure decomposition and the reverse load-pressure decomposition of the edge point is calculated as the positive and reverse load-pressure decomposition distance of the edge point; The load-pressure combined risk of the boundary point is obtained by combining the similarity of the forward and reverse order load-pressure decomposition quantities and the distance of the forward and reverse order load-pressure decomposition quantities of the boundary point.
2. The stress concentration data optimization method for high-pressure petrochemical containers according to claim 1 is characterized in that: in, In the plurality of different load steps, the starting load is non-zero, and the starting load corresponding to each mark point remains consistent, and the increment of the load step corresponding to each mark point remains consistent.
3. The stress concentration data optimization method for high-pressure petrochemical containers according to claim 1 is characterized in that: in, The three-dimensional model of the petrochemical pressure vessel is composed of a plurality of different coordinate points, and the three-dimensional model can be used to represent the three-dimensional geometric features of points, lines and surfaces of the petrochemical pressure vessel.
4. The method for optimizing stress concentration data of a high-pressure petrochemical container according to claim 1, characterized in that: in, The stress value of each boundary point corresponding to each load step is used to form the stress load vector of the boundary point. The stress load vector of each boundary point is decomposed along the positive order and reverse order of the increasing direction of the load step to obtain the positive order load-pressure decomposition and reverse order load-pressure decomposition, which are specifically: For each boundary point, an array consisting of stress values corresponding to each load step of the boundary point is used as the stress load vector of the boundary point; Among them, the stress load vector of the boundary point is decomposed along the positive order of the increasing direction of the load step to obtain the positive order load-pressure decomposition amount, specifically: traverse in the stress load vector, if the dimension before the current dimension is not empty, then for the stress value of the current dimension, calculate the maximum value of the stress values of each dimension before the current dimension as the positive order stress starting peak value of the current dimension, and take the ratio of the stress value of the current dimension to the positive order stress starting peak value of the current dimension as the positive order load-pressure decomposition value of the current dimension, thereby calculating the positive order load-pressure decomposition values of each dimension except the first dimension in the stress load vector, and take the array composed of the positive order load-pressure decomposition values of each dimension except the first dimension as the positive order load-pressure decomposition amount; Among them, the stress load vector of each boundary point is decomposed in the reverse order of the increasing direction of the load step to obtain the reverse load-pressure decomposition amount, specifically: traverse in the stress load vector, if the dimension after the current dimension is not empty, then for the stress value of the current dimension, the maximum value of the stress values of each dimension after the current dimension is calculated as the subsequent stress starting peak value of the current dimension, and the ratio of the stress value of the current dimension to the reverse stress starting peak value of the current dimension is used as the reverse load-pressure decomposition value of the current dimension, thereby calculating the reverse load-pressure decomposition values of each dimension except the first dimension in the stress load vector, and taking the array composed of the reverse load-pressure decomposition values of each dimension except the first dimension as the reverse load-pressure decomposition amount.
5. The method for optimizing stress concentration data of a high-pressure petrochemical container according to claim 1, characterized in that: in, According to the risk difference of the edge points in each edge, the edges with stress risk in the three-dimensional model are obtained, which are specifically: According to the risk amount difference of the edge marking points in each edge, rounded edge processing is performed on each edge respectively, wherein, if the risk amount difference of each edge marking point in an edge is greater than a threshold value, it is judged that the edge has stress risk under pressure conditions, and rounded edge processing is performed on the edge with stress risk in the three-dimensional model, and the threshold value is a value obtained by statistically analyzing the risk amount difference of each edge marking point.
6. A stress concentration data optimization system for high-pressure petrochemical containers, characterized in that: The stress concentration resistance data optimization system for high-pressure petrochemical containers runs on any computing device such as a desktop computer, a laptop computer or a cloud data center. The computing device includes: a processor, a memory and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps in the stress concentration resistance data optimization method for high-pressure petrochemical containers as described in any one of claims 1 to 5.
