Pipeline circumferential weld failure limit state prediction method, device, equipment and medium
Through the finite element model and random forest regression algorithm, the failure limit state prediction model of pipeline ring welds was established, which solved the problem that it is difficult to accurately predict the failure limit state of high-steel pipeline ring welds in the existing technology, and achieved high prediction accuracy and engineering application value.
Patent Information
- Application Number
- CN202411869968.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-18
- Publication Date
- 2025-05-16
AI Technical Summary
The existing technology lacks relatively accurate measures to predict the failure limit status of ring welds in high-steel pipelines, which makes it difficult to accurately determine the failure limit status of ring welds, affecting the pipe control, design and construction of newly built pipelines, as well as the safety assessment and maintenance decisions of in-service pipelines with defects.
Through the finite element model of the crack driving force of the pipeline ring weld, the crack driving force curve of the pipeline ring weld under different working conditions is obtained; the apparent fracture toughness is determined based on the crack resistance curve and passivation bias line; the true value of the limit strain capacity is determined based on the apparent fracture toughness and crack driving force curve; the failure limit state regression prediction formula is determined based on the wrong edge amount, wall thickness and internal pressure; the failure limit state regression prediction model is established using the random forest regression algorithm to further improve the prediction accuracy.
It realizes the ultimate strain ability of pipeline ring welds more accurately within the reasonable error range, improves the accuracy of the failure limit state of high-steel pipe ring welds, and supports more reliable design and maintenance decisions.
Smart Images
Figure CN120012471A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of engineering technology, and in particular to a method, device, equipment and medium for predicting the failure limit state of a pipeline girth weld. Background Art
[0002] Circumferential welding is an important part of the oil and gas pipeline construction process. During the welding process, the metal material is melted by the high temperature heat source, and then cooled and solidified to form a circumferential weld. The difference in the performance of the welding material and the pipe may lead to strength mismatch in the joint area, low strength matching of the circumferential weld joint and softening of the heat-affected zone. The welding process will inevitably lead to defects such as lack of fusion, pores, and cracks at the joint position. There are also a lot of joint misalignment and variable wall thickness caused by the circumferential welding process.
[0003] At present, the pipeline industry lacks relatively accurate means to predict the failure limit state of girth welds of high-grade steel pipelines, which makes it difficult to accurately determine the failure limit state of girth weld joints, affecting the pipe making, design and construction of new pipelines, as well as the safety assessment and maintenance and repair decisions of girth welds of defective in-service pipelines. Summary of the invention
[0004] The purpose of the embodiments of the present application is to provide a method, device, equipment and medium for predicting the failure limit state of a pipeline girth weld, so as to solve the technical defect that the prior art cannot relatively accurately predict the failure limit state of a high-grade steel pipeline girth weld.
[0005] The first aspect of the present application provides a pipeline girth weld failure limit state prediction method, comprising:
[0006] Through the crack driving force finite element model of pipeline girth weld, the crack driving force curve of pipeline girth weld under different working conditions is obtained;
[0007] Based on the crack resistance curve and passivation offset line of the pipeline girth weld, the apparent fracture toughness of the pipeline girth weld is determined;
[0008] Based on the apparent fracture toughness and crack driving force curve, the true value of the ultimate strain capacity of the pipeline girth weld is determined;
[0009] Determine the failure limit state regression prediction formula of the pipeline girth weld based on the misalignment, wall thickness and internal pressure;
[0010] Based on the failure limit state regression prediction formula of the pipeline girth weld, the first prediction value of the ultimate strain capacity of the pipeline girth weld is obtained;
[0011] According to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity, a first goodness of fit is obtained;
[0012] When the first goodness of fit is greater than the first preset threshold, it is determined that the failure limit state regression prediction formula is accurate.
[0013] In the embodiment of the present application, when the misalignment amount is 0, the wall thickness and the internal pressure are set values, the failure limit state regression prediction formula is:
[0014]
[0015] In the above formula, TSC Flu-1 represents the first predicted value of the ultimate strain capacity when the misalignment is 0, m represents the strength matching coefficient, yt represents the yield strength ratio, R represents the softening rate, c represents the crack length, a represents the crack depth, f1 represents the first intermediate transition coefficient, d1~d 14 Represents the formula fitting coefficient.
[0016] In the embodiment of the present application, when the misalignment amount is 0, the wall thickness and the internal pressure are respectively within the corresponding preset ranges, the failure limit state regression prediction formula is:
[0017]
[0018] In the above formula, TSC Flu-3 represents the first predicted value of the ultimate strain capacity when the misalignment is 0 and the wall thickness and internal pressure are within their corresponding preset ranges, t represents the wall thickness, p represents the internal pressure, and f represents the maximum strain capacity. t-1 represents the second intermediate action transition coefficient, y1 represents the third intermediate action transition coefficient, f p-1 represents the fourth intermediate action transition coefficient, y2 represents the fifth intermediate action transition coefficient, n1~n 12 is the formula fitting coefficient.
[0019] In the embodiment of the present application, when the misalignment amount is within a preset range, and the wall thickness and internal pressure are set values, the failure limit state regression prediction formula is:
[0020]
[0021] In the above formula, TSC Flu-2 represents the first predicted value of the ultimate strain capacity of the misalignment within the preset range, f2 represents the sixth intermediate action transition coefficient, f3 represents the seventh intermediate action transition coefficient, h represents the misalignment, g1~g 22 is the formula fitting coefficient.
[0022] In the embodiment of the present application, when the misalignment amount, wall thickness and internal pressure are respectively within the corresponding preset ranges, the failure limit state regression prediction formula is:
[0023]
[0024] In the above formula, TSC Flu-4 represents the first predicted value of the ultimate strain capacity of the misalignment, wall thickness and internal pressure within their corresponding preset ranges, f t-2 represents the eighth intermediate transition coefficient, f p-2 represents the ninth intermediate transition coefficient, y3 represents the tenth intermediate transition coefficient, y4 represents the eleventh intermediate transition coefficient, η1~η 14 is the formula fitting coefficient.
[0025] In an embodiment of the present application, the method further includes:
[0026] Based on the random forest regression algorithm, a neural network regression prediction model for the failure limit state of pipeline girth welds is established;
[0027] Inputting the working condition data into the failure limit state neural network regression prediction model, the second prediction value of the ultimate strain capacity of the pipeline girth weld is obtained;
[0028] According to the second predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity, a second goodness of fit is obtained;
[0029] When the second goodness of fit is greater than the second preset threshold, it is determined that the failure limit state neural network regression prediction model is accurate.
[0030] In an embodiment of the present application, the method further includes:
[0031] Obtain the validation set working condition data of pipeline girth weld;
[0032] Based on the full-scale pipeline numerical simulation model of the pipeline girth weld and the verification set working condition data, the verification value of the ultimate strain capacity of the pipeline girth weld is obtained;
[0033] Based on the failure limit state regression prediction formula and the validation set working condition data, the third prediction value of the ultimate strain capacity of the pipeline girth weld is obtained;
[0034] Input the validation set working condition data into the failure limit state neural network regression prediction model to obtain the fourth prediction value of the ultimate strain capacity of the pipeline girth weld;
[0035] A third relative error is obtained according to the ultimate strain capacity verification value and the third predicted value of the ultimate strain capacity;
[0036] A fourth relative error is obtained according to the ultimate strain capacity verification value and the fourth predicted value of the ultimate strain capacity;
[0037] When the difference between the third relative error and the fourth relative error is within a preset error range, it is determined that the failure limit state regression prediction formula is accurate.
[0038] A second aspect of the present application provides a pipeline girth weld failure limit state prediction device, comprising:
[0039] The first calculation module is used to obtain the crack driving force curve of the pipeline girth weld under different working condition data through the crack driving force finite element model of the pipeline girth weld;
[0040] A first determination module is used to determine the apparent fracture toughness of the pipeline girth weld based on the crack resistance curve and the passivation bias line of the pipeline girth weld;
[0041] The second determination module is used to determine the true value of the ultimate strain capacity of the pipeline girth weld based on the apparent fracture toughness and crack driving force curve;
[0042] The third determination module is used to determine the failure limit state regression prediction formula of the pipeline girth weld according to the misalignment amount, wall thickness and internal pressure;
[0043] A second calculation module is used to obtain a first prediction value of the ultimate strain capacity of the pipeline girth weld based on a failure limit state regression prediction formula of the pipeline girth weld;
[0044] A third calculation module is used to obtain a first goodness of fit according to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity;
[0045] The judgment module is used to determine that the failure limit state regression prediction formula is accurate when the first goodness of fit is greater than a first preset threshold.
[0046] The third aspect of the present application provides a pipeline girth weld failure limit state prediction device, comprising:
[0047] a memory configured to store instructions;
[0048] The processor is configured to call instructions from the memory and implement the pipeline girth weld failure limit state prediction method according to the first aspect when executing the instructions.
[0049] A fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run on a processor, the method for predicting the failure limit state of a pipeline girth weld as described in the first aspect above is executed.
[0050] The above technical solution establishes an initial failure limit state regression prediction formula by only considering the value of the misalignment, and further introduces wall thickness and internal pressure on the basis of the initial failure limit state regression prediction formula to obtain the final failure limit state regression prediction formula. The failure limit state regression prediction formula can be used to more accurately predict the ultimate strain capacity within a reasonable error range.
[0051] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:
[0053] Figure 1 Schematically showing a crack driving force curve and a crack resistance curve of a girth weld at different crack depths according to an embodiment of the present application;
[0054] Figure 2 A schematic diagram of a process flow of a pipeline girth weld failure limit state prediction method according to an embodiment of the present application is schematically shown;
[0055] Figure 3 A schematic diagram of a crack driving force curve of a pipeline girth weld according to an embodiment of the present application is schematically shown;
[0056] Figure 4 A schematic diagram of a crack resistance curve of a pipeline girth weld according to an embodiment of the present application is schematically shown;
[0057] Figure 5 A schematic diagram schematically shows a first predicted value of the ultimate strain capacity and a true value of the ultimate strain capacity according to an embodiment of the present application;
[0058] Figure 6 A schematic diagram schematically shows an ultimate strain capacity verification value, an ultimate strain capacity third prediction value, and an ultimate strain capacity fourth prediction value according to an embodiment of the present application. DETAILED DESCRIPTION
[0059] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0060] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.
[0061] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0062] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0063] In order to better illustrate the pipeline girth weld failure limit state prediction method provided in the embodiment of the present application, a brief description of the prior art is given.
[0064] The determination of the failure limit state of girth welds of high-grade steel pipelines is an important part of the structural design of girth welds and the assessment of engineering applicability. Foreign research institutions have preliminarily explored the variation law of girth weld strain capacity under the combined influence of multiple factors and established a girth weld ultimate strain capacity prediction model, but this prediction model cannot be accurately applied to the assessment of girth welds of high-grade steel pipelines in my country. Domestic related research has established a joint ultimate strain bearing capacity prediction model based on the actual situation of low-strength matching of welds in my country, but it has not accurately revealed the fracture failure mechanism of girth welds, and the accuracy of strain prediction results needs to be improved.
[0065] Figure 1 The following schematically shows the crack driving force curve and resistance curve of the girth weld at different crack depths according to the embodiment of the present application. Figure 1 As shown, Figure 1 (a) is the crack driving force curve at different crack depths, the horizontal axis represents the strain (mm / mm), and the vertical axis represents the crack tip opening displacement (mm). Figure 1(b) is the crack resistance curve at different crack depths. The horizontal axis represents the crack extension (mm), and the vertical axis represents the crack tip opening displacement (mm). Curve 10 represents a crack depth of 3.0 mm, curve 20 represents a crack depth of 4.0 mm, curve 30 represents a crack depth of 4.5 mm, curve 40 represents a crack depth of 6.0 mm, and curve 50 represents a crack depth of 5.0 mm. It can be seen that the crack driving force curve and the crack resistance curve at different crack depths are different. The increase in crack depth increases the crack driving force and reduces the apparent fracture toughness, so that the ultimate tensile stress variation of the pipeline girth weld is significantly reduced with the increase in crack depth. However, the prior art has not accurately quantified the apparent fracture toughness of the full-size pipeline girth weld, and has not taken into account that the crack driving force curve is not simply affected by a single factor, but is the result of the coupling of multiple factors, which has made it difficult to determine the failure limit state.
[0066] In addition, the critical values and correlations of joint material properties, defects and loads acting on the joints under limit states are still unclear, which affects the pipe manufacturing, design and construction of new pipelines and the safety assessment and maintenance and repair decisions of defective girth welds of in-service pipelines.
[0067] Based on the above technical defects, the embodiment of the present application proposes a method for predicting the failure limit state of a pipeline girth weld. When establishing a preliminary regression prediction formula for the failure limit state, first focus on the conditions where the misalignment is zero or the misalignment remains within a preset range. Then, on top of this basic formula, the two key factors of wall thickness and internal pressure are incorporated to improve the preliminary regression prediction formula for the failure limit state and obtain the final regression prediction formula for the failure limit state. The failure limit state regression prediction formula can achieve an accurate estimate of the ultimate strain capacity within a reasonable error range. It is worth mentioning that for the failure limit state of a pipeline girth weld, the failure limit state regression prediction formula shows a wide range of application potential. Among them, when the pipeline girth weld is subjected to external force, its strain capacity reaches its limit, and it will break or fail, thereby affecting the safe operation of the entire pipeline system, that is, the ultimate strain capacity characterizes the failure limit state of the pipeline girth weld.
[0068] Figure 2 The following schematically shows a flow chart of a pipeline girth weld failure limit state prediction method according to an embodiment of the present application. Figure 2 As shown, an embodiment of the present application provides a pipeline girth weld failure limit state prediction method, which may include the following steps:
[0069] S110: The crack driving force curve of the pipeline girth weld under different working condition data is obtained through the crack driving force finite element model of the pipeline girth weld.
[0070] Among them, the crack driving force finite element model is a tool that combines fracture mechanics and finite element analysis. It is used to simulate and analyze the crack propagation behavior and driving force in the pipeline girth weld. The working condition data are shown in Table 1. According to the different working condition data shown in Table 1, the crack driving force curve of the pipeline girth weld can be obtained.
[0071] Table 1
[0072]
[0073]
[0074] This embodiment takes the basic working condition data in Table 2 as an example to obtain the crack driving force curve of the pipeline girth weld under the basic working condition data. Figure 3 , Figure 3 The following schematically shows a crack driving force curve of a pipeline girth weld according to an embodiment of the present application. Figure 3 As shown, the horizontal axis represents strain (mm / mm) and the vertical axis represents crack tip opening displacement (mm).
[0075] Table 2
[0076]
[0077] S120: Determine the apparent fracture toughness of pipeline girth welds based on the crack resistance curve and passivation offset line of the pipeline girth welds.
[0078] See also Figure 4 , Figure 4 The following schematically shows a crack resistance curve diagram of a pipeline girth weld according to an embodiment of the present application. Figure 4 As shown, the horizontal axis represents the crack extension (mm), and the vertical axis represents the crack tip opening displacement (mm). The apparent fracture toughness of the two is determined based on the intersection of the 0.035mm passivation bias line with the fusion line crack and the center line crack. The apparent fracture toughness of the fusion line crack is 0.74mm, and the apparent fracture toughness of the center line crack is 1.33mm.
[0079] S130: Determine the true value of the ultimate strain capacity of the pipeline girth weld based on the apparent fracture toughness and crack driving force curves.
[0080] The apparent fracture toughness of the fusion line crack is 0.74 mm. Figure 3 The vertical axis corresponds to the intersection A (0.75%, 0.74) which is the initiation point of the fusion line crack of the girth weld. The true value of the ultimate strain capacity of the fusion line crack is 0.75%. The apparent fracture toughness of the centerline crack is 1.33 mm. Figure 3Corresponding to the vertical axis, the intersection B (1.31%, 1.33) is the initiation failure point of the girth weld of the centerline crack, and the true value of the ultimate strain capacity of the centerline crack is 1.31%.
[0081] S140: Determine the failure limit state regression prediction formula of the pipeline girth weld based on the misalignment, wall thickness and internal pressure.
[0082] Among them, in order to ensure the prediction accuracy of the failure limit state regression prediction formula, different failure limit state regression prediction formulas are used for different misalignment, wall thickness and internal pressure values. Therefore, it is necessary to determine the failure limit state regression prediction formula of the pipeline girth weld to be used.
[0083] S150: Based on the failure limit state regression prediction formula of the pipeline girth weld, a first prediction value of the ultimate strain capacity of the pipeline girth weld is obtained.
[0084] In this embodiment, based on the failure limit state regression prediction formula when the misalignment is 0 and the wall thickness and internal pressure are set values, formula correction coefficients are set to respectively give failure limit state regression prediction formulas when the wall thickness and internal pressure have other values.
[0085] When the misalignment is 0, the wall thickness and internal pressure are set values, the failure limit state regression prediction formula is:
[0086]
[0087] In the above formula, the wall thickness is 21.4mm, the internal pressure is 6MPa, and TSC Flu-1 represents the first predicted value of the ultimate strain capacity when the misalignment is 0, m represents the strength matching coefficient, yt represents the yield strength ratio, R represents the softening rate, c represents the crack length, a represents the crack depth, f1 represents the first intermediate transition coefficient, d1~d 14 Represents the formula fitting coefficient.
[0088] The above formula only involves the case where the misalignment is 0, and does not consider the influence of wall thickness and internal pressure on the ultimate strain capacity. The accuracy of this formula needs to be further improved.
[0089] In order to further improve the accuracy of the above formula, the influence of wall thickness and internal pressure on the ultimate strain capacity is considered, and the wall thickness and internal pressure are introduced into the following formula. When the misalignment is 0, the wall thickness and internal pressure are within the corresponding preset ranges, the failure limit state regression prediction formula is:
[0090]
[0091] In the above formula, the preset range of wall thickness is 21.4mm~30.8mm, the preset range of internal pressure is 0MPa~12Mpa, TSCFlu-3 represents the first predicted value of the ultimate strain capacity when the misalignment is 0 and the wall thickness and internal pressure are within their corresponding preset ranges, t represents the wall thickness, p represents the internal pressure, and f represents the maximum strain capacity. t-1 represents the second intermediate action transition coefficient, y1 represents the third intermediate action transition coefficient, f p-1 represents the fourth intermediate action transition coefficient, y2 represents the fifth intermediate action transition coefficient, n1~n 12 is the formula fitting coefficient.
[0092] When the misalignment is within the preset range and the wall thickness and internal pressure are set values, the failure limit state regression prediction formula is:
[0093]
[0094] In the above formula, the preset range of the misalignment is 0mm~3mm, the wall thickness is 21.4mm, the internal pressure is 6MPa, and TSC Flu-2 represents the first predicted value of the ultimate strain capacity of the misalignment within the preset range, f2 represents the sixth intermediate action transition coefficient, f3 represents the seventh intermediate action transition coefficient, h represents the misalignment, g1~g 22 is the formula fitting coefficient.
[0095] The above formula only involves the case where the misalignment is within the preset range, and does not consider the impact of wall thickness and internal pressure on the ultimate strain capacity. The accuracy of this formula needs to be further improved.
[0096] In order to further improve the accuracy of the above formula, the influence of wall thickness and internal pressure on the ultimate strain capacity is considered, and the wall thickness and internal pressure are introduced into the following formula. When the misalignment, wall thickness and internal pressure are within the corresponding preset ranges, the failure limit state regression prediction formula is:
[0097]
[0098] In the above formula, the preset range of misalignment is 0mm~3mm, the preset range of wall thickness is 21.4mm~30.8mm, the preset range of internal pressure is 0MPa~12Mpa, TSC Flu-4 represents the first predicted value of the ultimate strain capacity of the misalignment, wall thickness and internal pressure within their corresponding preset ranges, f t-2 represents the eighth intermediate transition coefficient, f p-2 represents the ninth intermediate transition coefficient, y3 represents the tenth intermediate transition coefficient, y4 represents the eleventh intermediate transition coefficient, η1~η 14 is the formula fitting coefficient.
[0099] This embodiment takes into account the strong coupling relationship between the misalignment amount and the strength matching coefficient and crack size parameter of the girth weld, and therefore expresses the failure limit state regression prediction formula of the pipeline girth weld with and without misalignment in the form of a piecewise expression.
[0100] S160: Obtain a first goodness of fit according to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity.
[0101] See also Figure 5 , Figure 5 The schematic diagram schematically shows the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity according to the embodiment of the present application. Figure 5 As shown, Figure 5 The horizontal axis of (a) represents the true value of the ultimate strain capacity, and the vertical axis represents the first predicted value of the ultimate strain capacity when the misalignment is 0 and the wall thickness and the internal pressure are within the corresponding preset ranges. Figure 5 The horizontal axis of (b) represents the true value of the ultimate strain capacity, and the vertical axis represents the first predicted value of the ultimate strain capacity when the misalignment, wall thickness and internal pressure are within the corresponding preset ranges. Among them, the goodness of fit is an important indicator to measure the degree of consistency between the predicted value and the true value. The closer the goodness of fit is to 1, the more accurate the predicted value is. For the case where the misalignment is 0, the goodness of fit of the failure limit state regression prediction formula is 0.88, and for the case where the misalignment is within the preset range, the goodness of fit of the failure limit state regression prediction formula is 0.86.
[0102] S170: When the first goodness of fit is greater than a first preset threshold, determining that the failure limit state regression prediction formula is accurate.
[0103] Among them, for the case where the misalignment amount is 0, the goodness of fit of the failure limit state regression prediction formula is 0.88, and for the case where the misalignment amount is within the preset range, the goodness of fit of the failure limit state regression prediction formula is 0.86, indicating that the failure limit state regression prediction formula with a misalignment amount of 0 and the failure limit state regression prediction formula with a misalignment amount within the preset range both have high accuracy. The first preset threshold is set according to actual needs and is not specifically limited in this embodiment.
[0104] It is worth noting that this embodiment is developed based on the girth weld joint of the X80 pipeline. Considering that a single change in yield strength has little effect on the apparent fracture toughness and ultimate strain capacity of the girth weld, this embodiment is also applicable to girth weld joints of other high-grade steel pipelines.
[0105] In an optional implementation, the method further includes:
[0106] S210: Based on the random forest regression algorithm, a neural network regression prediction model for the failure limit state of pipeline girth welds is established.
[0107] Among them, the failure limit state neural network regression prediction model of this embodiment is an existing neural network model. On the basis of the existing failure limit state neural network regression prediction model, a random forest regression algorithm is introduced. The basic component unit of the random forest is a decision tree. The decision tree is a feature-based recursive segmentation method that performs predictions by gradually dividing the features. Each internal node represents a feature, and each leaf node represents a prediction result. Random forest introduces the concept of randomness to improve the generalization ability of the model. When constructing each decision tree, a training set is created by sampling with replacement from the training data, and only a randomly selected part of the features is considered at each node for division. Random forest trains multiple decision trees and integrates their prediction results by taking the average of all decision trees to obtain more stable and accurate predictions.
[0108] S220: Inputting the operating condition data into the failure limit state neural network regression prediction model to obtain a second prediction value of the ultimate strain capacity of the pipeline girth weld.
[0109] Among them, multiple samples are randomly selected from the working condition data with replacement through a random sampling method to generate multiple sub-datasets. For each sub-dataset, a regression tree is constructed using the decision tree algorithm. When each node is split, a part of the features is randomly selected, and the best feature is selected for splitting. The overfitting problem of the decision tree algorithm is avoided, and the engineering applicability of the random forest regression method is improved. After all trees are trained, each decision tree is used to predict the newly input data points, and then all prediction results are averaged to obtain the second prediction value of the ultimate strain capacity.
[0110] S230: Obtain a second goodness of fit according to the second predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity.
[0111] Among them, the second goodness of fit of the failure limit state neural network regression prediction model obtained is greater than 0.99.
[0112] S240: When the second goodness of fit is greater than a second preset threshold, determining that the failure limit state neural network regression prediction model is accurate.
[0113] Among them, when the second goodness of fit is greater than 0.99, it means that the accuracy of the failure limit state neural network regression prediction model is relatively high. The second preset threshold is set according to actual needs and is not specifically limited in this embodiment.
[0114] In an optional implementation, the method further includes:
[0115] S310: Obtain verification set working condition data of the pipeline girth weld.
[0116] Among them, the verification set working condition data of the pipeline girth weld are shown in Table 3. The load form is tensile and the crack location is the fusion line crack.
[0117] Table 3
[0118]
[0119] S320: Based on the full-scale pipeline numerical simulation model of the pipeline girth weld and the verification set working condition data, the ultimate strain capacity verification value of the pipeline girth weld is obtained.
[0120] Among them, the full-size pipeline numerical simulation model of the pipeline girth weld is a three-dimensional geometric model of the full-size pipeline established by the three-dimensional modeling software according to the actual size of the pipeline and the geometric characteristics of the girth weld, reflecting the actual pipeline and weld structure to obtain the ultimate strain capacity verification value.
[0121] S330: Based on the failure limit state regression prediction formula and the verification set working condition data, a third prediction value of the ultimate strain capacity of the pipeline girth weld is obtained.
[0122] S340: Input the verification set working condition data into the failure limit state neural network regression prediction model to obtain a fourth prediction value of the ultimate strain capacity of the pipeline girth weld.
[0123] Among them, see Figure 6 , Figure 6 The schematic diagram schematically shows the ultimate strain capacity verification value, the ultimate strain capacity third prediction value and the ultimate strain capacity fourth prediction value according to the embodiment of the present application. Figure 6 As shown, the horizontal axis represents the ultimate strain capacity verification value, the vertical axis represents the predicted value, the solid line represents that the relative error between the predicted value and the ultimate strain capacity verification value is 0, and the dotted line represents that the relative errors between the predicted value and the ultimate strain capacity verification value are +10% and -10%, respectively.
[0124] S350: Obtain a third relative error according to the ultimate strain capacity verification value and the third ultimate strain capacity prediction value.
[0125] Among them, the third relative error=(the third predicted value of the ultimate strain capacity-the verification value of the ultimate strain capacity) / the verification value of the ultimate strain capacity×100%, and it can be obtained that the maximum value of the third relative error is 17.2%.
[0126] S360: Obtain a fourth relative error according to the ultimate strain capacity verification value and the fourth predicted value of the ultimate strain capacity.
[0127] Among them, the fourth relative error=(the fourth predicted value of the ultimate strain capacity-the ultimate strain capacity verification value) / the ultimate strain capacity verification value×100%, and it can be obtained that the maximum value of the fourth relative error is less than 15%.
[0128] S370: When the difference between the third relative error and the fourth relative error is within a preset error range, determining that the failure limit state regression prediction formula is accurate.
[0129] Among them, overall, the fourth relative error between the fourth predicted value of the ultimate strain capacity predicted by the failure limit state neural network regression prediction model and the ultimate strain capacity verification value is less than 15%. Although the third relative error between the third predicted value of the ultimate strain capacity of the failure limit state regression prediction formula and the ultimate strain capacity verification value is greater than the fourth relative error between the fourth predicted value of the ultimate strain capacity predicted by the failure limit state neural network regression prediction model and the ultimate strain capacity verification value, the third relative error is still within an acceptable and reasonable range. The accuracy of the failure limit state regression prediction formula of this embodiment meets the requirements of engineering applications. The preset error range is set according to actual needs, and this embodiment does not make specific limitations.
[0130] Optionally, the embodiment of the present application further provides a pipeline girth weld failure limit state prediction device, comprising:
[0131] The first calculation module is used to obtain the crack driving force curve of the pipeline girth weld under different working condition data through the crack driving force finite element model of the pipeline girth weld;
[0132] A first determination module is used to determine the apparent fracture toughness of the pipeline girth weld based on the crack resistance curve and the passivation bias line of the pipeline girth weld;
[0133] The second determination module is used to determine the true value of the ultimate strain capacity of the pipeline girth weld based on the apparent fracture toughness and crack driving force curve;
[0134] The third determination module is used to determine the failure limit state regression prediction formula of the pipeline girth weld according to the misalignment amount, wall thickness and internal pressure;
[0135] A second calculation module is used to obtain a first prediction value of the ultimate strain capacity of the pipeline girth weld based on a failure limit state regression prediction formula of the pipeline girth weld;
[0136] A third calculation module is used to obtain a first goodness of fit according to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity;
[0137] The judgment module is used to determine that the failure limit state regression prediction formula is accurate when the first goodness of fit is greater than a first preset threshold.
[0138] Optionally, the embodiment of the present application further provides a pipeline girth weld failure limit state prediction device, comprising:
[0139] a memory configured to store instructions;
[0140] The processor is configured to call instructions from the memory and implement the pipeline girth weld failure limit state prediction method according to any one of the above items when executing the instructions.
[0141] Optionally, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run on a processor, the method for predicting the failure limit state of a pipeline girth weld as described above is executed.
[0142] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0143] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0144] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0146] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0147] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0148] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0149] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0150] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A pipeline girth weld failure limit state prediction method, characterized in that: include: By using the crack driving force finite element model of the pipeline girth weld, a crack driving force curve of the pipeline girth weld under different working condition data is obtained; Determining the apparent fracture toughness of the pipeline girth weld based on the crack resistance curve and the passivation bias line of the pipeline girth weld; Determining a true value of the ultimate strain capacity of the pipeline girth weld based on the apparent fracture toughness and the crack driving force curve; Determine a failure limit state regression prediction formula of the pipeline girth weld according to the misalignment amount, wall thickness and internal pressure; Based on the failure limit state regression prediction formula of the pipeline girth weld, a first prediction value of the ultimate strain capacity of the pipeline girth weld is obtained; Obtaining a first goodness of fit according to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity; When the first goodness of fit is greater than a first preset threshold, it is determined that the failure limit state regression prediction formula is accurate.
2. The method according to claim 1, characterized in that When the misalignment amount is 0, the wall thickness and the internal pressure are set values, the failure limit state regression prediction formula is: In the above formula, TSC Flu-1 represents the first predicted value of the ultimate strain capacity when the misalignment is 0, m represents the strength matching coefficient, yt represents the yield strength ratio, R represents the softening rate, c represents the crack length, a represents the crack depth, f1 represents the first intermediate transition effect coefficient, d1~d 14 Represents the formula fitting coefficient.
3. The method according to claim 1, characterized in that When the misalignment amount is 0, and the wall thickness and the internal pressure are respectively within the corresponding preset ranges, the failure limit state regression prediction formula is: In the above formula, TSC Flu-3 represents the first predicted value of the ultimate strain capacity when the misalignment is 0 and the wall thickness and the internal pressure are respectively within their corresponding preset ranges, t represents the wall thickness, p represents the internal pressure, and f ... wall thickness and the internal pressure are respectively within their corresponding preset ranges, t-1 represents the second intermediate action transition coefficient, y1 represents the third intermediate action transition coefficient, f p-1 represents the fourth intermediate action transition coefficient, y2 represents the fifth intermediate action transition coefficient, n1~n 12 is the formula fitting coefficient.
4. The method according to claim 1, characterized in that: When the misalignment amount is within a preset range, and the wall thickness and the internal pressure are set values, the failure limit state regression prediction formula is: In the above formula, TSC Flu-2 represents the first predicted value of the ultimate strain capacity of the misalignment within a preset range, f2 represents the sixth intermediate action transition coefficient, f3 represents the seventh intermediate action transition coefficient, h represents the misalignment, g1~g 22 is the formula fitting coefficient.
5. The method according to claim 1, characterized in that When the misalignment amount, the wall thickness and the internal pressure are respectively within the corresponding preset ranges, the failure limit state regression prediction formula is: In the above formula, TSC Flu-4 represents the first predicted value of the ultimate strain capacity of the misalignment amount, the wall thickness and the internal pressure within their corresponding preset ranges, f t-2 represents the eighth intermediate transition coefficient, f p-2 represents the ninth intermediate transition coefficient, y3 represents the tenth intermediate transition coefficient, y4 represents the eleventh intermediate transition coefficient, η1~η 14 is the formula fitting coefficient.
6. The method according to claim 1, characterized in that The method further comprises: According to the random forest regression algorithm, a neural network regression prediction model for the failure limit state of the pipeline girth weld is established; Inputting the operating condition data into the failure limit state neural network regression prediction model to obtain a second prediction value of the ultimate strain capacity of the pipeline girth weld; Obtaining a second goodness of fit according to the second predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity; When the second goodness of fit is greater than a second preset threshold, it is determined that the failure limit state neural network regression prediction model is accurate.
7. The method according to claim 6, characterized in that The method further comprises: Obtaining verification set working condition data of the pipeline girth weld; Based on the full-scale pipeline numerical simulation model of the pipeline girth weld and the verification set working condition data, obtaining the ultimate strain capacity verification value of the pipeline girth weld; Based on the failure limit state regression prediction formula and the verification set working condition data, a third prediction value of the ultimate strain capacity of the pipeline girth weld is obtained; Inputting the validation set working condition data into the failure limit state neural network regression prediction model to obtain a fourth prediction value of the ultimate strain capacity of the pipeline girth weld; Obtaining a third relative error according to the ultimate strain capacity verification value and the third predicted value of the ultimate strain capacity; Obtaining a fourth relative error according to the ultimate strain capacity verification value and the fourth predicted value of the ultimate strain capacity; In the case where the difference between the third relative error and the fourth relative error is within a preset error range, it is determined that the failure limit state regression prediction formula is accurate.
8. A pipeline girth weld failure limit state prediction device, characterized in that: include: A first calculation module is used to obtain a crack driving force curve of the pipeline girth weld under different working condition data through a crack driving force finite element model of the pipeline girth weld; A first determination module, configured to determine the apparent fracture toughness of the pipeline girth weld based on a crack resistance curve and a passivation bias line of the pipeline girth weld; A second determination module is used to determine a true value of the ultimate strain capacity of the pipeline girth weld based on the apparent fracture toughness and the crack driving force curve; A third determination module is used to determine a failure limit state regression prediction formula of the pipeline girth weld according to the misalignment amount, wall thickness and internal pressure; A second calculation module is used to obtain a first prediction value of the ultimate strain capacity of the pipeline girth weld based on a failure limit state regression prediction formula of the pipeline girth weld; A third calculation module, used for obtaining a first goodness of fit according to the first predicted value of the ultimate strain capacity and the true value of the ultimate strain capacity; A judgment module is used to determine whether the failure limit state regression prediction formula is accurate when the first goodness of fit is greater than a first preset threshold.
9. A pipeline girth weld failure limit state prediction device, characterized in that: include: a memory configured to store instructions; A processor is configured to call the instructions from the memory and implement the pipeline girth weld failure limit state prediction method according to any one of claims 1-7 when executing the instructions.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is run on a processor, the method for predicting the failure limit state of a pipeline girth weld according to any one of claims 1 to 7 is executed.