A method and device for predicting failure pressure of a pipeline containing defects and a computer device

Abstract

The present application provides a kind of defective pipeline failure pressure prediction method, device and computer equipment, comprising: obtaining the pipeline parameter of defective pipeline, the defect parameter at defect and the external force parameter of the external force suffered by the defective pipeline, the external force parameter includes the axial force and bending moment suffered by the defect;The pipeline parameter, the defect parameter and the external force parameter are input into the pre-trained failure pressure prediction model to determine the corresponding failure pressure of fluid when the defective pipeline reaches failure strength;The failure pressure prediction model is trained according to a plurality of pipeline parameters, a plurality of defect parameters, a plurality of external force parameters and corresponding failure pressure, can realize comprehensive characterization of pipeline and each parameter at defect, and obtain more accurate failure pressure of pipeline, to guide pipeline management personnel to set reasonable operating pressure, avoid pipeline rupture failure.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas pipeline safety, and particularly relates to a failure pressure prediction method and device for a pipeline with defects and a computer device. BACKGROUND

[0002] With the rapid development of economy, the demand for energy is increasing year by year, and pipelines with large capacity have become the primary choice for oil and gas transportation in China. At the same time, the geological conditions and corrosion environment faced by oil and gas pipelines during operation are becoming more and more complex, which makes the failure pressure evaluation of defective pipelines under external load gradually become an important part of pipeline integrity management and reliability evaluation. Therefore, it is necessary to accurately and efficiently determine the failure pressure of the pipeline with defects under complex load, and set a reasonable operating pressure according to the failure pressure to ensure the safety and efficiency of oil and gas transportation.

[0003] When evaluating the limit failure pressure of a single defect pipeline, the existing evaluation methods all take the ultimate tensile strength of the pipeline material as the failure criterion when the defect pipeline occurs strength failure. However, when calculating the limit failure pressure of the pipeline, the existing calculation methods only consider the pipeline parameters of the pipeline with defects and the defect parameters at the defect, which leads to a large difference between the theoretical calculation value and the actual value of the limit failure pressure of the pipeline with defects, misleading pipeline management personnel to use a larger internal pressure, increasing the possibility of causing pipeline rupture failure. SUMMARY

[0004] In view of the above problems of the prior art, the purpose of the present application is to provide a failure pressure prediction method and device for a pipeline with defects and a computer device to solve the problem that the prior art only considers the pipeline parameters of the pipeline with defects and the defect parameters at the defect, leading to a large difference between the theoretical value and the actual value of the limit failure pressure of the pipeline with defects, misleading pipeline management personnel to use a larger internal pressure, and causing pipeline rupture.

[0005] To solve the above technical problems, the specific technical solutions of the present application are as follows:

[0006] On the one hand, the present application provides a failure pressure prediction method for a pipeline with defects, comprising:

[0007] obtaining pipeline parameters of a pipeline with defects, defect parameters at the defect, and external force parameters of the external force acting on the pipeline with defects, the external force parameters including the axial force and bending moment acting on the defect;

[0008] inputting the pipeline parameter, the defect parameter and the external force parameter into a pre-trained failure pressure prediction model to determine a failure pressure corresponding to the fluid when the defective pipeline reaches a failure strength; the failure pressure prediction model is trained according to a plurality of pipeline parameters, a plurality of defect parameters, a plurality of external force parameters and corresponding failure pressures.

[0009] As an embodiment herein, the failure strength includes a yield failure strength, a flow failure strength and an ultimate failure strength;

[0010] The failure pressure includes a yield failure internal pressure corresponding to the yield failure strength, a flow failure internal pressure corresponding to the flow failure strength and an ultimate failure internal pressure corresponding to the ultimate failure strength.

[0011] As an embodiment herein, the pre-training method of the failure pressure prediction model includes:

[0012] obtaining a data set composed of a plurality of pipeline parameters of defective pipelines, a plurality of defect parameters at defect positions on the defective pipelines and a plurality of external force parameters of the defective pipelines;

[0013] inputting the data set into an adjusted state simulation model to obtain a plurality of failure pressures corresponding one-to-one to target pipeline parameters, target defect parameters and target external force parameters in the data set; wherein the boundary conditions of the state simulation model are adjusted according to a burst pressure test experiment of the defective pipeline;

[0014] inputting a plurality of pipeline parameters, a plurality of defect parameters, a plurality of external force parameters and a plurality of failure pressures calculated by the state simulation model into a back propagation neural network for training to obtain the failure pressure prediction model taking the pipeline parameter, the defect parameter and the external force parameter as input and taking the failure pressure as target output.

[0015] As an embodiment herein, the data set composed of a plurality of pipeline parameters of defective pipelines, a plurality of defect parameters at defect positions on the defective pipelines and a plurality of external force parameters of the defective pipelines further includes:

[0016] determining a first value range of the pipeline parameter according to the American Petroleum Institute standard, and randomly generating the pipeline parameter of the defective pipeline within the first value range;

[0017] determining a second value range of the defect parameter and the external force parameter according to the pipeline parameter, and randomly generating the defect parameter and the external force parameter of the defective pipeline within the second value range;

[0018] The dataset consists of several randomly generated pipeline parameters, several defect parameters, and several external force parameters.

[0019] As an embodiment of this document, the step of determining a first range of values ​​for the pipeline parameters according to American Petroleum Institute standards, and randomly generating the pipeline parameters for the defective pipeline within the first range of values, further includes:

[0020] The pipes are classified according to the pipe material and yield strength in the American Petroleum Institute standards, and in each classification, the first range of values ​​for the pipe parameters of the defective pipes is determined.

[0021] The pipe parameters of the defective pipe are randomly generated within the first range of values ​​for the pipe parameters of this type of pipe.

[0022] As an embodiment of this paper, the step of importing the dataset into the adjusted state simulation model to obtain several failure pressures that correspond one-to-one with the target pipeline parameters, target defect parameters, and target external force parameters in the dataset further includes:

[0023] The dataset is imported into a state simulation model to construct a virtual pipeline with defects.

[0024] Internal pressure is continuously applied to the defective virtual pipeline, and the triaxial stress at the defect of the virtual pipeline is determined by comparing it with the target external force parameters in the dataset.

[0025] The comprehensive equivalent stress at the defect is determined based on the triaxial stress.

[0026] When the combined equivalent stress reaches the three failure intensities of the defective pipeline, the three internal pressures applied to the defective virtual pipeline are recorded, and the three internal pressure values ​​are recorded as the failure pressure of the defective pipeline.

[0027] As an embodiment of this document, the step of continuously applying internal pressure to the defective virtual pipeline and determining the equivalent triaxial stress at the defect location of the virtual pipeline using the target external force parameters in the dataset further includes:

[0028] Apply full constraints to one end node of the defective virtual pipe;

[0029] Set a reference point at the center position of the other end of the pipe and apply an external force according to the target external force parameters;

[0030] The pipe node on the other end of the pipe is connected to the reference point using a coupling technique, so that the external force applied to the reference point is applied to the defective virtual pipe.

[0031] The circumferential, radial, and axial stresses at the defect on the other end of the pipe are used as the calculated values ​​of the equivalent stress.

[0032] As one embodiment of this article, the triaxial stress includes circumferential stress, radial stress, and axial stress;

[0033] The step of determining the equivalent stress at the defect based on the triaxial stress further includes:

[0034] According to the formula:

[0035]

[0036] The equivalent stress σ is obtained, where σ h The circumferential stress is σ. r The radial stress is σ. l The axial stress is mentioned above.

[0037] On the other hand, this paper also provides a failure pressure prediction device for defective pipelines, including:

[0038] The acquisition unit is used to acquire the pipeline parameters of the defective pipeline, the defect parameters at the defect, and the external force parameters of the external forces acting on the defective pipeline, wherein the external force parameters include the axial force and bending moment acting on the defect.

[0039] The prediction unit is used to input the pipeline parameters, the defect parameters, and the external force parameters into a pre-trained failure pressure prediction model to determine the failure pressure of the fluid when the defective pipeline reaches the failure strength; the failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure.

[0040] On the other hand, this document also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned method for predicting the failure pressure of a defective pipeline.

[0041] By employing the above technical solution, by acquiring the pipeline parameters of the defective pipeline, the defect parameters at the defect location, and the external force parameters of the defective pipeline, including the axial force and bending moment at the defect location, it is possible to obtain pipeline parameters and defect parameters that characterize the pipeline and defect characteristics. It is also possible to acquire the external force parameters of the region where the pipeline is located, including the axial force and bending moment at the defect location, thus providing a more comprehensive characterization of the various parameters of the pipeline and the defect location. By inputting the pipeline parameters, defect parameters, and external force parameters into a pre-trained failure pressure prediction model, the failure pressure corresponding to the fluid when the defective pipeline reaches its failure strength can be determined. The failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure. This allows for a more accurate acquisition of the pipeline's failure pressure based on a more comprehensive characterization of the various parameters of the pipeline and the defect location, ensuring that the theoretical and actual values ​​of the ultimate failure pressure of the defective pipeline differ minimally. This, in turn, guides pipeline managers to adopt reasonable internal pressure to prevent pipeline rupture.

[0042] To make the above and other objects, features and advantages of this document more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments or prior art described herein, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this article. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 An overall system diagram of a failure pressure prediction method for a defective pipeline, as described in this embodiment, is shown.

[0045] Figure 2 A schematic diagram illustrating the steps of a failure pressure prediction method for a defective pipeline according to an embodiment of this paper is shown.

[0046] Figure 3 A schematic diagram of the training method for the failure pressure prediction model in the embodiments of this paper is shown;

[0047] Figure 4 The following diagrams illustrate the defects of the embodiments described in this article before and after simplification.

[0048] Figure 5 A scatter plot of the first random parameter distribution in the embodiments of this paper is shown;

[0049] Figure 6A scatter plot of the second random parameter distribution in the embodiments of this paper is shown;

[0050] Figure 7 A scatter plot of the random parameter distribution in the third embodiment of this paper is shown;

[0051] Figure 8 This diagram illustrates the mesh partitioning of the state simulation model in the embodiments described in this paper.

[0052] Figure 9 This diagram illustrates a method for obtaining a dataset using a state simulation model, as illustrated in the embodiments of this paper.

[0053] Figure 10 This diagram illustrates the applied loads and constraints in the state simulation model of the embodiments described in this paper.

[0054] Figure 11 The network structure of the failure stress prediction model in the embodiments of this paper is shown;

[0055] Figure 12 The figure shows a graph of burst test data for a single defect pipeline in the embodiments described in this article;

[0056] Figure 13 This document presents a detailed comparison chart of ultimate failure pressures in the embodiments described herein.

[0057] Figure 14 A scatter plot of comparative data for several models in the embodiments of this paper is shown.

[0058] Figure 15 A schematic diagram of a failure pressure prediction device for a defective pipeline, as described in this embodiment, is shown.

[0059] Figure 16 A schematic diagram of a computer device as described in this article is shown.

[0060] Explanation of symbols in the attached drawings:

[0061] 101. Terminal;

[0062] 102. Computing server;

[0063] 1501, Acquisition Unit;

[0064] 1502, Prediction Unit;

[0065] 1602. Computer equipment;

[0066] 1604, Processor;

[0067] 1606. Memory;

[0068] 1608. Drive mechanism;

[0069] 1610. Input / Output Module;

[0070] 1612. Input devices;

[0071] 1614. Output devices;

[0072] 1616. Presentation equipment;

[0073] 1618. Graphical User Interface;

[0074] 1620. Network interface;

[0075] 1622. Communication link;

[0076] 1624. Communication bus. Detailed Implementation

[0077] The technical solutions in the embodiments described below will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments described herein, and not all of the embodiments. Based on the embodiments described herein, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this document.

[0078] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings herein are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0079] like Figure 1 The diagram shows an overall system diagram of a failure pressure prediction method for a defective pipeline, including: a terminal 101 and a computing server 102.

[0080] Terminal 101 is used to receive pipe parameters, defect parameters, and external force parameters of the pipeline to be predicted input by the user. The external force parameters can be obtained from sensors on the pipeline, the defect parameters can be obtained through geometric calculations, and the pipe parameters can be obtained from the user manual. The terminal sends the received pipe parameters, defect parameters, and external force parameters to the computing server. In this paper, terminal 101 can be wirelessly or wiredly connected to computing server 102.

[0081] The computing server 102 is used to run the failure pressure prediction model and input the received pipeline parameters, defect parameters and external force parameters into the failure pressure prediction model. The failure pressure of the pipeline to be predicted is calculated by the failure pressure prediction model and the terminal 101 is prevented. The failure pressure prediction model in this paper takes pipeline parameters, defect parameters and external force parameters as inputs and outputs yield failure internal pressure, flow failure internal pressure and ultimate failure internal pressure.

[0082] Through repeated research, the inventors have determined that pipelines typically endure complex external loads during actual operation, such as soil settlement, landslides, earthquakes, and frost heave. Axial and bending loads caused by these factors are the most common forms of external loads on pipelines. Furthermore, complex operating conditions can lead to deep degradation phenomena such as cracks, manufacturing defects, and inclusions at pipeline defects. These phenomena can cause pipelines to rupture before reaching their ultimate failure pressure, resulting in safety issues. Therefore, existing technologies that rely solely on pipeline parameters, defect parameters at the defect location, and the ultimate tensile strength criterion to determine the failure pressure of defective pipelines are overly simplistic.

[0083] To address the aforementioned issues, this paper presents a method for predicting the failure pressure of defective pipelines. This method comprehensively considers numerous factors influencing pipeline failure pressure, thereby obtaining a failure pressure that closely approximates the actual failure pressure of the pipeline. Furthermore, it increases the selectivity for pipeline managers in assessing the failure pressure of defective pipelines. Figure 2 This is a schematic diagram illustrating the steps of a failure pressure prediction method for a defective pipeline provided in this embodiment. This specification provides the operational steps of the method described in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or device products, the methods shown in the embodiments or drawings can be executed sequentially or in parallel. Specifically, as shown in the attached diagrams... Figure 2 As shown, the method may include:

[0084] Step 201: Obtain the pipe parameters of the defective pipe, the defect parameters at the defect, and the external force parameters of the defective pipe, wherein the external force parameters include the axial force and bending moment at the defect.

[0085] Step 202: Input the pipeline parameters, the defect parameters, and the external force parameters into the pre-trained failure pressure prediction model to determine the failure pressure of the fluid when the defective pipeline reaches the failure strength; the failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure.

[0086] By employing the above technical solution, by acquiring the pipeline parameters of the defective pipeline, the defect parameters at the defect location, and the external force parameters of the external forces acting on the defective pipeline, including the axial force and bending moment at the defect location, it is possible to obtain pipeline parameters and defect parameters that characterize the pipeline characteristics and defect characteristics. It is also possible to acquire the external force parameters of the region where the pipeline is located, including the axial force and bending moment at the defect location, thus providing a more comprehensive characterization of various parameters of the pipeline and the defect location. By inputting the pipeline parameters, defect parameters, and external force parameters into a pre-trained failure pressure prediction model, the failure pressure corresponding to the fluid when the defective pipeline reaches its failure strength can be determined. The failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure. This allows for a more accurate acquisition of the pipeline's failure pressure based on a more comprehensive characterization of various parameters of the pipeline and the defect location, ensuring that the theoretical and actual values ​​of the ultimate failure pressure of the defective pipeline differ minimally. This, in turn, guides pipeline managers in setting reasonable operating pressures to prevent pipeline rupture failure.

[0087] As one embodiment of this article, the failure strength includes yield failure strength, flow failure strength, and ultimate failure strength;

[0088] The failure pressure includes the yield failure internal pressure corresponding to the yield failure strength, the flow failure internal pressure corresponding to the flow failure strength, and the ultimate failure internal pressure corresponding to the ultimate failure strength.

[0089] In this step, compared to existing technologies that can only obtain the ultimate failure internal pressure of a defective pipeline, this paper can also obtain the yield failure internal pressure and the flow failure internal pressure of a defective pipeline. Therefore, it can more comprehensively characterize the internal pressure of the pipeline in different states, giving pipeline managers more options. The yield failure internal pressure is less than the flow failure internal pressure, and the flow failure internal pressure is less than the ultimate failure internal pressure.

[0090] In this paper, three strength failure criteria (characterized by three material strengths) correspond to three failure internal pressures. The three material strengths are yield strength, flow strength (between yield and ultimate tensile strength) and ultimate tensile strength. As the internal pressure of the pipeline increases, the equivalent stress at the pipeline defect will first reach the yield stress, then the flow stress, and finally the ultimate tensile stress (strength is stress).

[0091] In this paper, the yield, flow, and ultimate tensile strength of materials are defined because steel materials undergo elastic deformation first under tensile stress. As the stress increases, the material first enters the yield state, which can be understood as the material changing from elastic deformation to plastic deformation. Then, the material undergoes plastic flow deformation. Flow strength is an artificial definition, not an intrinsic property of the material. It is generally defined as the average of the yield and ultimate tensile strength (there are other definitions). Finally, the material no longer has sufficient plastic resistance to tensile deformation and has basically fractured at the ultimate tensile strength, breaking apart and no longer having the ability to coordinate deformation.

[0092] like Figure 3 The diagram illustrates the training method for the failure pressure prediction model. In this paper, the failure pressure prediction model is trained using several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressures. Specifically, the training method for the failure pressure prediction model includes:

[0093] Step 301: Obtain a dataset consisting of several pipeline parameters of defective pipelines, several defect parameters of defects at defects on the defective pipelines, and several external force parameters of external forces acting on the defective pipelines.

[0094] In this step, the dataset contains multiple sets of data. Each set of data includes the pipeline parameters of a defective pipeline, the defect parameters of the defective pipeline, and the external force parameters of the defective pipeline under the current environment. In this paper, the sources of external forces on the defective pipeline include soil settlement, landslides, earthquakes, and soil freezing.

[0095] In this paper, the defects on the defective pipeline may be formed due to external corrosion, and the shape of the defects may be irregular. To simplify the subsequent calculation process, before inputting the dataset into the state simulation model, the irregularly shaped defects can be simplified to regular shapes, such as... Figure 4 The diagram shows the defects before and after simplification. Figure 4 In this context, defects can be simplified to rectangles. When recording defect parameters, only the length L, width W, and thickness d of the simplified rectangle need to be recorded. Furthermore, Figure 4 In this diagram, D represents the diameter of the defective pipe, and t represents the wall thickness of the defective pipe. This method can reduce the recording time for defect recorders and improve the efficiency of defect recording.

[0096] Specifically, in this step,

[0097] The first range of values ​​for the pipeline parameters is determined according to the American Petroleum Institute standard, and the pipeline parameters for the defective pipeline are randomly generated within the first range of values.

[0098] In this step, pipe parameters are generated by randomly selecting values ​​based on the range of values ​​for pipe outer diameter, wall thickness, yield strength, and tensile strength of X42-X80 pipe steel specified in API SPECIFICATION 5L (American Petroleum Institute Standard).

[0099] The second range of values ​​for the defect parameters and the external force parameters is determined based on the pipeline parameters, and the defect parameters and the external force parameters of the pipeline containing the defect are randomly generated within the second range of values.

[0100] In this step, based on the randomly generated pipe parameters, and following the conditions of 0.2D≤L≤1.4D, 2t≤w≤14t, 0.1t≤d≤0.8t, and -0.4F... ref ≤F≤0, 0≤M≤0.4M ref The rules randomly generate defect parameters and external force parameters. In this paper, defect parameters include defect length, defect width, and defect depth. External force parameters include axial force and bending moment parameters.

[0101] To prevent mismatches in parameters such as pipe diameter and wall thickness caused by an excessively large range of random parameters, this paper can also classify the pipes according to the pipe material and yield strength in the American Petroleum Institute standard, and determine the first range of pipe parameters for each category containing defects.

[0102] The pipe parameters of the defective pipe are randomly generated within the first range of values ​​for the pipe parameters of this type of pipe.

[0103] In this paper, using the X60 pipeline as a boundary, two large groups are set up: X42-X60 and X60-X80, to ensure parameter randomization while preventing excessive value mismatch. The obtained random parameters are input into a validated state simulation model for calculation, ultimately yielding 354 sets of failure pressure data for X42-X60 pipelines, 234 sets for X60-X80 pipelines, and 100 sets for X42-X80 pipelines under internal pressure load only. The random parameters and simulation results essentially cover most types of currently operational oil and gas pipelines. Therefore, the data in this paper is comprehensive enough to be used for training subsequent prediction models.

[0104] The dataset consists of several randomly generated pipeline parameters, several defect parameters, and several external force parameters.

[0105] like Figure 5 — Figure 7The scatter plot of the random parameter distribution shown includes pipe diameter ranging from 219.34 to 963.68 mm, pipe wall thickness ranging from 9.01 to 19.99 mm, yield strength ranging from 290.1 ​​to 705 MPa, tensile strength ranging from 400.3 to 855.6 MPa, defect length ranging from 50.45 to 1300.24 mm, defect width ranging from 18.95 to 228.91 mm, defect depth ranging from 1.023 to 15.465 mm, axial compressive stress ranging from -9599.294 to 0 kN, and closed bending moment ranging from 0 to 1278.826 kN·m, basically covering most types of oil and gas pipelines currently in service.

[0106] This step can obtain the dataset using numerical simulation methods or physical simulation experiments. This article does not specify which method is preferred.

[0107] Step 302: Input the dataset into the adjusted state simulation model to obtain several failure pressures that correspond one-to-one with the target pipeline parameters, target defect parameters, and target external force parameters in the dataset; wherein, the boundary conditions of the state simulation model are adjusted according to the burst pressure test experiment of the defective pipeline.

[0108] In this step, the state simulation model is first established:

[0109] A three-dimensional state simulation model of a defective pipeline under the combined action of internal pressure, axial force, and bending moment was established using the APDL programming language of the finite element software ANSYS. By programming the state simulation model, parameters can be easily and quickly modified, enabling rapid modeling under different parameters.

[0110] like Figure 8 The diagram shows the mesh generation of the state simulation model. In the finite element software, the 20-node hexahedral element (Solid186) was selected to construct the main body of the defective pipe due to its excellent performance in simulating large plastic deformation. In this paper, considering both stress concentration at the defect location under external forces and computational efficiency, the mesh at the defect location was refined, while the mesh far from the defect location was thinned to ensure good simulation results while reducing computation time.

[0111] The state simulation model established in this paper was verified by selecting 14 sets of burst pressure test data from published literature. These 14 sets of data included three steel grades: X46, X52 and X60, and three different load conditions: internal pressure only, combined internal pressure and axial pressure, and combined internal pressure and bending moment. These data can well represent the complex load problems involved in this invention. Detailed information is shown in Table 1, which is a comparison table of the actual burst test and the ultimate burst pressure of the finite element simulation.

[0112] Table 1. Comparison of Ultimate Burst Pressure between Actual Burst Tests and Finite Element Simulations

[0113]

[0114] The comparison results show that, regardless of the load condition, the established state simulation model can obtain good simulation results under the corresponding parameters of the actual blasting test, proving that the settings in the state simulation model, including the mesh, boundary conditions, and load application methods, are reasonable and reliable. Therefore, using this state simulation model to study the failure pressure of defective pipelines under varying parameters has high reliability and can provide convenient and reliable database information for the training and verification of artificial neural networks.

[0115] After the state simulation model is established, such as Figure 9 The diagram shows a method for obtaining a dataset using a state simulation model.

[0116] Step 901: Import the dataset into the state simulation model to construct a virtual pipeline with defects;

[0117] Step 902: Continuously apply internal pressure to the virtual pipeline containing defects, and determine the triaxial stress at the defect of the virtual pipeline by comparing it with the target external force parameters in the dataset;

[0118] Step 903: Determine the comprehensive equivalent stress at the defect based on the triaxial stress;

[0119] Step 904: When the comprehensive equivalent stress reaches the three failure intensities of the defective pipeline, record the three internal pressures applied to the defective virtual pipeline, and record the three internal pressure values ​​as the failure pressure of the defective pipeline.

[0120] As an example of this article, such as Figure 10 The schematic diagram of the applied load and constraint conditions in the state simulation model shown, step 904, which involves continuously applying internal pressure to the virtual pipeline with defects and determining the equivalent triaxial stress at the defect location of the virtual pipeline using the target external force parameters in the dataset, further includes:

[0121] Apply full constraints to one end node of the defective virtual pipe;

[0122] Set a reference point at the center position of the other end of the pipe and apply an external force according to the target external force parameters;

[0123] The pipe node on the other end of the pipe is connected to the reference point using a coupling technique, so that the external force applied to the reference point is applied to the defective virtual pipe.

[0124] The circumferential, radial, and axial stresses at the defect on the other end of the pipe are used as the calculated values ​​of the equivalent stress.

[0125] In this finite element simulation, all nodes on section A (outer interface) are fully constrained, while section B (inner section) is free. A reference point C is set at the center of section B to apply axial load F and bending load M. A coupling technique is used to connect the pipe nodes on the inner section to the reference point C, allowing the mechanical forces applied at reference point C to be transferred to the pipe elements on the inner section through the connection characteristics. This simulates the external forces acting on the pipe, where P represents the internal pressure, F represents the axial force, and M represents the bending moment. Since the ultimate goal of the finite element simulation is to determine the failure pressure of a defective pipe, when applying loads, the axial force F and bending moment M are first applied simultaneously at reference point C. Then, a monotonically increasing internal pressure load is applied to the inner surface of the pipe until pipe failure occurs.

[0126] The determination of the comprehensive equivalent stress at the defect based on the triaxial stress further includes:

[0127] According to the formula:

[0128]

[0129] The equivalent stress σ is obtained, where σ h The circumferential stress is σ. r The radial stress is σ. l The axial stress is mentioned above.

[0130] When the combined equivalent stress reaches the three failure intensities of the defective pipeline, the three internal pressures applied to the defective virtual pipeline are recorded, and the three internal pressures are recorded as the failure pressures of the defective pipeline, further including:

[0131] When the comprehensive equivalent stress reaches the yield failure strength of the defective pipeline, the internal pressure applied to the defective virtual pipeline is recorded, and this internal pressure is recorded as the yield failure internal pressure of the defective pipeline.

[0132] When the comprehensive equivalent stress reaches the flow failure intensity of the defective pipeline, the internal pressure applied to the defective virtual pipeline is recorded, and this internal pressure is recorded as the flow failure internal pressure of the defective pipeline.

[0133] When the comprehensive equivalent stress reaches the ultimate failure strength of the defective pipeline, the internal pressure applied to the defective virtual pipeline is recorded, and this internal pressure is recorded as the ultimate failure internal pressure of the defective pipeline.

[0134] Step 303: Import several pipeline parameters, several defect parameters, several external force parameters, and several failure pressures calculated by the state simulation model as a training set into the backpropagation neural network for training, to obtain the failure pressure prediction model with pipeline parameters, the defect parameters, and the external force parameters as inputs and the failure pressure as the target output.

[0135] In selecting hyperparameters for the failure stress prediction model, the choice of activation function and hidden layer structure (number of layers and number of nodes per layer) is particularly important. Eight commonly used activation functions (Sigmoid function, Swish function, ReLU function, LeakyReLU function, ELU function, Tanh function, Softmax function, and Softplus function) were substituted into the initially constructed backpropagation neural network model. The backpropagation neural network model was trained using 688 sets of data obtained through a state simulation model. By observing the changes in the training set loss and validation set loss values ​​of the eight activation functions during the training process, the ELU activation function, which exhibits fast convergence speed, relatively close training set loss values ​​to validation set loss values, and small fluctuations, was ultimately selected as the activation function for the failure stress prediction model in this method.

[0136] After determining the activation function, the following parameters were kept constant: ELU activation function, 10% test set selection rate for validation, 2200 iterations, and 156 sets of data per training iteration. Only the number of hidden layers and nodes in each layer were changed. The number of hidden layers was varied by 1 at intervals from 2 to 11, and the number of nodes was varied by 20 at intervals from 20 to 200 for each hidden layer. The hidden layer structure was evaluated by comparing the validation set loss during model training. Ultimately, a hidden layer structure with 7 layers and 180 nodes per layer was determined to have the best prediction performance. Since the failure stress prediction model under the current hyperparameter settings has reached a low level in both training and test set losses, it can be considered that the settings of other hyperparameters are basically within the optimal range and will not be discussed in detail.

[0137] like Figure 11 The network structure of the failure pressure prediction model shown is presented. Thus, a complete failure pressure prediction model capable of predicting the failure pressure of various pipelines and pipeline defects under complex loads has been established. The input layer of this failure pressure prediction model has 9 input vectors: pipeline outer diameter, wall thickness, yield strength, tensile strength, defect length, defect width, defect depth, axial force, and bending moment. The hidden layers are set to 7 layers, each with 180 nodes. The output layer has 3 output vectors: ultimate failure pressure, flow failure pressure, and yield failure pressure.

[0138] After the failure pressure prediction model is established, the accuracy of the prediction model can be verified. The specific process includes model self-verification and model external verification.

[0139] Model self-validation:

[0140] In the self-validation phase, the basic strategy was as follows: 673 sets of data were randomly selected from a total sample database of 688 sets as training data, and the remaining 15 sets were used as validation data. To eliminate the influence of subjective factors, random sampling was used for both training and validation data. A total of 6 sets of self-validation work for the failure pressure prediction model were conducted. For each validation, 673 sets of training data and 15 sets of validation data were randomly selected from the total database. Accuracy verification was performed on ultimate failure pressure, flow failure pressure, and yield failure pressure, respectively. A total of 90 sets of prediction results and FEA results were compared. The calculated average relative errors of the 90 sets of failure pressure prediction results were 2.23%, 2.14%, and 3.12%, respectively. This indicates that the established failure pressure prediction model has high reliability in both accuracy and stability, and can basically replace the finite element method to quickly obtain failure pressure data of defective pipelines under conditions of pipeline outer diameter, wall thickness, yield strength, tensile strength, defect length, defect width, defect depth, axial force, and bending moment.

[0141] External model validation:

[0142] In the external verification phase, the established failure pressure prediction model was used to verify the actual blasting test data collected from the literature. Since the defective steel pipes used in the actual blasting tests were basically manufactured under strict machining conditions, the defects would generally not exhibit deterioration failure pressures such as cracks. Therefore, when using the failure pressure prediction model for prediction, the ultimate failure pressure results from the failure pressure analysis were compared with those of the two models.

[0143] like Figure 12The diagram shows the burst test data of a single-defect pipeline. 27 sets of data were collected (sorted by diameter from smallest to largest). Simultaneously, several commonly used empirical formulas for predicting ultimate failure pressure were selected to predict the failure pressure of these 27 sets of burst tests. The results were then compared with the predictions from the established failure pressure prediction model. Figure 13 The detailed comparison of ultimate failure pressures is shown in the diagram. It should be noted that the BPNN model established in this invention can not only predict the conventional ultimate failure internal pressure of actual defective pipelines, but also simultaneously and rapidly predict the corresponding flow failure internal pressure and yield failure internal pressure. Specific data are shown in columns 3 and 4 of Table 3. Figure 14 The scatter plots comparing the data from several models show that the B31G, Mod B31G, and Shell 92 models have poor prediction accuracy in these 27 sets of data. In terms of average relative error, the BPNN model in this method has the smallest average relative error at only 7.45%, while the Shell 92, Mod B31G, and B31G models all have average relative errors exceeding 10%, with a maximum of 19.11%. Although the PCORRC and CUP models have relatively high accuracy, neither can predict the ultimate failure internal pressure conditions of defective pipelines under external loads.

[0144] like Figure 15 The schematic diagram shown is of a failure pressure prediction device for a defective pipeline, comprising:

[0145] The acquisition unit 1501 is used to acquire the pipeline parameters of the defective pipeline, the defect parameters at the defect, and the external force parameters of the external forces acting on the defective pipeline, wherein the external force parameters include the axial force and bending moment acting on the defect.

[0146] The prediction unit 1502 is used to input the pipeline parameters, the defect parameters and the external force parameters into a pre-trained failure pressure prediction model to determine the failure pressure of the fluid when the defective pipeline reaches the failure strength; the failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters and the corresponding failure pressure.

[0147] By employing the above technical solution, the acquisition unit can acquire pipeline parameters and defect parameters that characterize pipeline features and defect characteristics. It can also acquire external force parameters of the region where the pipeline is located, including the axial force and bending moment at the defect. This allows for a more comprehensive characterization of various parameters of the pipeline and the defect. The prediction unit, based on this more comprehensive characterization of pipeline and defect parameters, can accurately acquire the pipeline's failure pressure, ensuring that the theoretical and actual values ​​of the ultimate failure pressure of the defective pipeline differ minimally. This guides pipeline management personnel to adopt appropriate internal pressure to prevent pipeline rupture.

[0148] like Figure 16 As shown in this embodiment, a computer device 1602 may include one or more processors 1604, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The computer device 1602 may also include any memory 1606 for storing information of any kind, such as code, settings, data, etc. Non-limitingly, for example, the memory 1606 may include any type of RAM, any type of ROM, flash memory, hard disk, optical disk, etc. More generally, any memory can use any technology to store information. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 1602. In one case, when the processor 1604 executes associated instructions stored in any memory or combination of memories, the computer device 1602 may perform any operation of the associated instructions. The computer device 1602 also includes one or more drive mechanisms 1608 for interacting with any memory, such as hard disk drive mechanisms, optical disk drive mechanisms, etc.

[0149] Computer device 1602 may also include an input / output module 1610 (I / O) for receiving various inputs (via input device 1612) and providing various outputs (via output device 1614). A specific output mechanism may include a presentation device 1616 and an associated graphical user interface (GUI) 1618. In other embodiments, the input / output module 1610 (I / O), input device 1612, and output device 1614 may be omitted, and the device may function solely as a computer device within a network. Computer device 1602 may also include one or more network interfaces 1620 for exchanging data with other devices via one or more communication links 1622. One or more communication buses 1624 couple the components described above together.

[0150] Communication link 1622 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. Communication link 1622 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0151] Corresponding to Figures 2 to 14 In addition to the methods described above, this embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the above-described methods.

[0152] This embodiment also provides a computer-readable instruction, wherein when a processor executes the instruction, the program therein causes the processor to perform the following: Figures 2 to 14 The method shown.

[0153] It should be understood that in the various embodiments of this document, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0154] It should also be understood that, in the embodiments herein, the term "and / or" is merely a description of the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following associated objects have an "or" relationship.

[0155] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this document.

[0156] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0157] In the embodiments provided herein, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or units, or they may be electrical, mechanical, or other forms of connection.

[0158] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments described herein, depending on actual needs.

[0159] Furthermore, the functional units in the various embodiments of this document can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0160] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this paper, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this paper. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0161] This document uses specific embodiments to illustrate the principles and implementation methods of this document. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this document. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this document. Therefore, the content of this specification should not be construed as a limitation of this document.

Claims

1. A method for predicting the failure pressure of a defective pipeline, characterized in that, include: Obtain the pipeline parameters of the defective pipeline, the defect parameters at the defect location, and the external force parameters of the external forces acting on the defective pipeline, wherein the external force parameters include the axial force and bending moment acting at the defect location; The pipeline parameters, defect parameters, and external force parameters are input into a pre-trained failure pressure prediction model to determine the failure pressure of the fluid when the defective pipeline reaches its failure strength; the failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure. The failure strength includes yield failure strength, flow failure strength, and ultimate failure strength; The failure pressure includes the yield failure internal pressure corresponding to the yield failure strength, the flow failure internal pressure corresponding to the flow failure strength, and the ultimate failure internal pressure corresponding to the ultimate failure strength. The pre-training method for the failure pressure prediction model includes: Obtain a dataset consisting of several pipe parameters containing defects, several defect parameters at the defects on the pipes containing defects, and several external force parameters of the external forces acting on the pipes containing defects. The dataset is input into the adjusted state simulation model to obtain several failure pressures that correspond one-to-one with the target pipeline parameters, target defect parameters, and target external force parameters in the dataset; wherein, the boundary conditions of the state simulation model are adjusted based on the burst pressure test experiment of the defective pipeline; Several pipeline parameters, several defect parameters, several external force parameters, and several failure pressures calculated by the state simulation model are used as a training set and imported into the backpropagation neural network for training to obtain the failure pressure prediction model with the pipeline parameters, the defect parameters, and the external force parameters as inputs and the failure pressure as the target output.

2. The failure pressure prediction method for defective pipelines according to claim 1, characterized in that, The process of obtaining a dataset consisting of several pipe parameters of defective pipes, several defect parameters at defects on the defective pipes, and several external force parameters of external forces acting on the defective pipes further includes: The first range of values ​​for the pipeline parameters is determined according to the American Petroleum Institute standard, and the pipeline parameters for the defective pipeline are randomly generated within the first range of values. The second range of values ​​for the defect parameters and the external force parameters is determined based on the pipeline parameters, and the defect parameters and the external force parameters of the pipeline containing the defect are randomly generated within the second range of values. The dataset consists of several randomly generated pipeline parameters, several defect parameters, and several external force parameters.

3. The failure pressure prediction method for defective pipelines according to claim 2, characterized in that, The step of determining a first range of values ​​for the pipeline parameters according to American Petroleum Institute standards, and randomly generating the pipeline parameters for the defective pipeline within the first range of values, further includes: The pipes are classified according to the pipe material and yield strength in the American Petroleum Institute standards, and in each classification, the first range of values ​​for the pipe parameters of the defective pipes is determined. The pipe parameters of the defective pipe are randomly generated within the first range of values ​​for the pipe parameters of this type of pipe.

4. The failure pressure prediction method for defective pipelines according to claim 1, characterized in that, The step of importing the dataset into the adjusted state simulation model to obtain several failure pressures that correspond one-to-one with the target pipeline parameters, target defect parameters, and target external force parameters in the dataset further includes: The dataset is imported into a state simulation model to construct a virtual pipeline with defects. Internal pressure is continuously applied to the defective virtual pipeline, and the triaxial stress at the defect of the virtual pipeline is determined by comparing it with the target external force parameters in the dataset. The comprehensive equivalent stress at the defect is determined based on the triaxial stress. When the combined equivalent stress reaches the three failure intensities of the defective pipeline, the three internal pressures applied to the defective virtual pipeline are recorded, and the three internal pressure values ​​are recorded as the failure pressure of the defective pipeline.

5. The failure pressure prediction method for defective pipelines according to claim 4, characterized in that, The step of continuously applying internal pressure to the defective virtual pipeline and determining the equivalent triaxial stress at the defect location of the virtual pipeline using the target external force parameters in the dataset further includes: Apply full constraints to one end node of the defective virtual pipe; Set a reference point at the center of the other end of the pipe and apply an external force according to the target external force parameters; The pipe node on the other end of the pipe is connected to the reference point using a coupling technique, so that the external force applied to the reference point is applied to the defective virtual pipe. The circumferential, radial, and axial stresses at the defect on the other end of the pipe are used as the calculated values ​​of the equivalent stress.

6. The failure pressure prediction method for defective pipelines according to claim 4, characterized in that, The triaxial stress includes circumferential stress, radial stress, and axial stress; The step of determining the equivalent stress at the defect based on the triaxial stress further includes: According to the formula: The equivalent stress σ is obtained, where σ h The circumferential stress is σ. r The radial stress is σ. l The axial stress is mentioned above.

7. A failure pressure prediction device for a defective pipeline, characterized in that, include: The acquisition unit is used to acquire the pipeline parameters of the defective pipeline, the defect parameters at the defect, and the external force parameters of the external forces acting on the defective pipeline, wherein the external force parameters include the axial force and bending moment acting on the defect. The prediction unit is used to input the pipeline parameters, the defect parameters, and the external force parameters into a pre-trained failure pressure prediction model to determine the failure pressure of the fluid when the defective pipeline reaches the failure strength; the failure pressure prediction model is trained based on several pipeline parameters, several defect parameters, several external force parameters, and the corresponding failure pressure. The failure strength includes yield failure strength, flow failure strength, and ultimate failure strength; The failure pressure includes the yield failure internal pressure corresponding to the yield failure strength, the flow failure internal pressure corresponding to the flow failure strength, and the ultimate failure internal pressure corresponding to the ultimate failure strength. The pre-training method for the failure pressure prediction model includes: acquiring a dataset consisting of several pipeline parameters of defective pipelines, several defect parameters at defects on the defective pipelines, and several external force parameters of the external forces acting on the defective pipelines; inputting the dataset into a adjusted state simulation model to obtain several failure pressures that correspond one-to-one with the target pipeline parameters, target defect parameters, and target external force parameters in the dataset; wherein the boundary conditions of the state simulation model are adjusted based on the burst pressure test experiment of the defective pipelines; and importing the several pipeline parameters, several defect parameters, several external force parameters, and several failure pressures calculated by the state simulation model as a training set into a backpropagation neural network for training to obtain the failure pressure prediction model with the pipeline parameters, the defect parameters, and the external force parameters as inputs and the failure pressure as the target output.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the failure pressure prediction method for defective pipelines as described in any one of claims 1-6.

Images