Multi-defect pipeline failure pressure evaluation method and system

Abstract

The invention relates to the technical field of safety assessment, in particular to a multi-defect pipeline failure pressure assessment method and system, and the method comprises the steps: calculating failure pressure based on different pipelines and defect parameters, and building a prediction model; determining defect parameters according to the detection data of the pipeline, respectively determining an axial projection line and a circumferential projection line through a target defect center in sequence, and respectively projecting defects in a projection area along the axial direction and the circumferential direction of the pipeline; the defects on the projection lines are overlapped, and calculating the parameters of the equivalent single defect according to the effective area of the overlapped defects; the equivalent single defects on the projection lines interact with one another, and parameters of the equivalent single defects which do not affect one another are calculated according to the effective area of the interaction defects; all the equivalent single defect parameters which do not affect each other and the corresponding pipeline parameters are used for obtaining the failure pressure of the target pipeline through the prediction model. By considering the axial and circumferential distribution influence of the defects and the interaction between the defects, the problem of excessive simplification is solved, and good accuracy and universality are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety assessment, and in particular to a method and system for assessing failure pressure of a multi-defect pipeline. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] During the long-term service of pipelines, due to factors such as corrosion and erosion, material loss will occur, resulting in volume defects, which will reduce the pressure-bearing capacity of the pipeline. Once the pipeline fails, it will not only affect the fluid transportation, but also cause safety accidents. Defective pipeline assessment is carried out based on pipeline inspection data to determine the failure pressure of defective pipelines, thereby reflecting the pressure limit of the pipeline and providing a theoretical basis for the safe operation and effective maintenance of the pipeline.

[0004] Pipeline defects are often complex in shape and randomly distributed. In order to reduce the complexity of failure pressure assessment of multi-defect pipelines, existing methods usually simplify the defect shape or simplify the interaction of multiple defects. For example, multiple defects are simplified into single defects or double defects with regular shapes, or the influence of circumferential distribution of defects is ignored.

[0005] Prior art CN116576402A proposes a method for predicting the failure pressure of irregularly shaped defective pipelines. Based on the characteristics of the depth profile of irregularly shaped defects, the shape parameters of irregular defects are calculated, and the defects are divided into three equivalent shapes, namely rectangular, parabolic and mixed. Based on the effective depth and evaluation length of the equivalent shape, the failure pressure of irregularly shaped defective pipelines is evaluated. However, this method does not consider the influence of the interaction of multiple defects, cannot accurately evaluate the failure pressure of multi-defective pipelines, and has high requirements for defect profile data, and the prediction and evaluation efficiency is low.

[0006] Prior art CN117131727A proposes a method for determining the failure pressure of a pipeline containing group corrosion defects, which simplifies multiple defects into two most obvious defects, trains an interaction parameter prediction model based on a failure data set, predicts interaction parameters based on pipeline parameters and defect parameters, and determines the failure pressure of the pipeline under the combined effect of multiple defects. However, when more than two defects on the target pipeline have an influence that cannot be ignored, this method oversimplifies the mutual influence of multiple defects and reduces the evaluation accuracy.

[0007] However, when there are significant interactions between defects, this simplification is often excessive and will lead to reduced evaluation accuracy. Especially when the circumferential distribution of defects is more obvious, the circumferential impact of defects cannot be ignored. Therefore, the existing evaluation methods are conservative in their calculation results and it is difficult to accurately evaluate the failure pressure of multi-defect pipelines. In addition, there are problems such as low evaluation efficiency and limited scope of application. Summary of the invention

[0008] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method and system for assessing the failure pressure of a multi-defect pipeline, which comprehensively considers the influence of the axial and circumferential distribution of defects, forms a multi-defect equivalent simplification method, combines a neural network algorithm with the multi-defect equivalent simplification, and has good accuracy, high efficiency and versatility.

[0009] In order to achieve the above object, the present invention adopts the following technical solution: A first aspect of the present invention provides a method for assessing failure pressure of a multi-defect pipeline, comprising the following steps: Based on the finite element model of a single defect pipeline with different pipeline and defect parameters, the failure pressure is calculated to form a failure data training set. By training the neural network, a failure pressure prediction model based on the neural network is established; According to the inspection data of the pipeline, the defect parameters on the pipeline are determined, the target defects are selected in turn, the axial projection line and the circumferential projection line are determined through the center of the target defect, the projection area is determined, and the defects in the projection area are projected along the axial direction and circumferential direction of the pipeline respectively; If there is overlap between defects on the projection line, the parameters of the equivalent single defect are calculated based on the effective area of ​​the overlapping defects; If there is interaction between the equivalent single defects on the projection line, the parameters of the equivalent single defects are calculated according to the effective area of ​​the interaction defects, and the interaction defects are simplified into equivalent single defects that do not affect each other; All the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on neural network to obtain the failure pressure of the target pipeline.

[0010] Furthermore, a failure pressure prediction model based on a neural network is established, including: using a single-defect pipeline finite element model with different pipelines and different defect parameters to calculate the corresponding failure pressure, and the obtained failure pressure forms a failure data training set for training the neural network.

[0011] Furthermore, establishing a failure pressure prediction model based on a neural network also includes: using pipeline parameters and defect parameters of a failure data training set as input data and the failure pressure of a single defective pipeline as output data, training a neural network, and establishing a failure pressure prediction model based on a neural network.

[0012] Furthermore, according to the inspection data of the pipeline, the defect parameters on the pipeline are determined, the target defects are selected in turn, the axial projection line and the circumferential projection line are determined through the center of the target defect, and the projection area is determined, and the defects in the projection area are projected along the axial direction and circumferential direction of the pipeline respectively; specifically: Determine the size and location parameters of the defects on the pipeline, select the target defects in turn, determine the axial and circumferential projection lines through the center of the target defect, and form a projection area with two projection lines parallel to the corresponding projection lines and with a set spacing. Project the defects in the projection area onto the axial and circumferential projection lines; the set spacing is the spacing between the axial and circumferential projection lines, as shown in the following formula: ; ; in, Z is the spacing between axial projection lines, S is the spacing between circumferential projection lines, and Represent the axial and circumferential interaction distances, respectively.

[0013] Furthermore, according to the effective area of ​​the overlapping defects, the parameters of the equivalent single defect are calculated as shown in the following formula: ; ; ;

[0014] in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the effective length of the defect in the overlapping area, is the effective width of the defect in the overlap region, is the effective area of ​​the defect axial section in the overlap region, is the effective area of ​​the defect radial section in the overlapping area.

[0015] Furthermore, there is interaction between equivalent single defects, specifically: the spacing between equivalent single defects is smaller than the corresponding axial or circumferential interaction spacing.

[0016] Furthermore, according to the effective area of ​​the interaction defect, the parameters of the equivalent single defect are calculated as shown in the following formula: ; ; ; in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the interaction defect length, is the interaction defect width, is the effective area of ​​the axial section of the interaction defect, is the effective area of ​​the radial section of the combined defect, is the spacing between interacting defects.

[0017] Furthermore, all the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on the neural network to obtain the failure pressure of the target pipeline; specifically: All the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on neural network in turn, and the minimum value is selected as the failure pressure at the target defect. The influence range of the defect failure pressure is determined based on the defect center position and the interaction distance. When the ranges overlap, the minimum failure pressure value is selected to obtain the target pipeline failure pressure.

[0018] A second aspect of the present invention provides a multi-defect pipeline failure pressure assessment system, comprising: The failure pressure prediction unit is configured to: calculate the failure pressure based on the single defect pipeline finite element model with different pipeline and defect parameters, form a failure data training set, train the neural network, and establish a failure pressure prediction model based on the neural network; The projection simplification unit is configured to: determine the defect parameters on the pipeline according to the inspection data of the pipeline, select the target defects in sequence, determine the axial projection line and the circumferential projection line through the center of the target defect respectively, and determine the projection area, and project the defects in the projection area along the axial direction and circumferential direction of the pipeline respectively; The multi-defect pipeline analysis unit is configured to: if the number of defects on the projection line exceeds one and there is overlap between the defects, then the parameters of the equivalent single defect are calculated according to the effective area of ​​the overlapping defects, and the irregular overlapping defects are simplified into regular equivalent single defects; The multi-defect pipeline analysis unit is further configured to: if there is interaction between equivalent single defects on the projection line, calculate the parameters of the equivalent single defects according to the effective area of ​​the interacting defects, and simplify the interacting defects into equivalent single defects that do not affect each other; The failure pressure prediction unit is further configured to: input all the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other into the failure pressure prediction model based on the neural network to obtain the failure pressure of the target pipeline.

[0019] Compared with the prior art, one or more of the above technical solutions have the following beneficial effects: 1. Compared with the traditional evaluation method that oversimplifies the interaction of multiple defects, especially the impact of the circumferential distribution of defects, this scheme comprehensively considers the impact of defects in the axial and circumferential distribution, simplifies overlapping defects and interacting defects, and finally obtains equivalent single defects that do not affect each other. The equivalent single defect parameters and pipeline parameters are then input into the pre-trained neural network model to calculate the failure pressure of the target pipeline, which significantly improves the evaluation accuracy of multi-defect pipelines.

[0020] 2. Compared with the traditional assessment method that relies too much on experience to evaluate the failure pressure of multi-defect pipelines, this solution combines the neural network algorithm with multi-defect equivalent simplification, avoids manual intervention, efficiently processes pipeline inspection data, and effectively improves the assessment efficiency of multi-defect pipelines.

[0021] 3. Compared with the relatively single evaluation results of traditional evaluation methods, the target pipeline failure pressure in this scheme includes the failure pressure of each defect and the influence range of the failure pressure, which presents the failure pressure of the defective pipeline more vividly and helps on-site personnel to carry out pipeline maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0023] Figure 1 is a schematic diagram of a failure pressure assessment process for a multi-defect pipeline provided by one or more embodiments of the present invention; Figure 2 is a schematic diagram of a simplified process of multi-defect equivalence provided by one or more embodiments of the present invention; Figure 3 is a schematic diagram of axial defect projection and overlapping defect processing provided by one or more embodiments of the present invention; Figure 4 is a schematic diagram of circumferential defect projection and overlapping defect processing provided by one or more embodiments of the present invention; Figure 5 It is a simplified schematic diagram of an interaction defect equivalent provided by one or more embodiments of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0025] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0026] As introduced in the background technology, the existing assessment methods are somewhat conservative in their calculation results, making it difficult to accurately assess the failure pressure of multi-defect pipelines, which may lead to unnecessary maintenance and pipe replacement. Therefore, a new method for assessing the failure pressure of multi-defect pipelines is proposed. Based on pipeline inspection data, the accuracy of the assessment of the failure pressure of multi-defect pipelines is improved through equivalent simplification of multiple defects and accurate prediction of neural networks.

[0027] Embodiment 1: The following embodiment provides a method for assessing the failure pressure of a multi-defect pipeline. Based on a finite element model of a single-defect pipeline with different pipelines and defect parameters, the failure pressure is calculated to form a failure data training set, a neural network is trained, and a failure pressure prediction model based on a neural network is established. The pipeline inspection data is obtained, and multiple defects are simplified into equivalent single defects that do not affect each other by axial and circumferential projections of the defects, as well as equivalent simplifications of overlapping defects and interacting defects. The equivalent single defect parameters that do not affect each other and the corresponding pipeline parameters are input into the failure pressure prediction model based on the neural network to obtain the target pipeline failure pressure. Compared with the existing methods, this method comprehensively considers the influence of the axial and circumferential distribution of defects, combines the neural network algorithm with the equivalent simplification of multiple defects, and has good accuracy, high efficiency and versatility.

[0028] The specific implementation process is as follows: (1) Based on the finite element model of a single-defect pipeline with different pipeline and defect parameters, the failure pressure is calculated to form a failure data training set, train the neural network, and establish a failure pressure prediction model based on the neural network; (2) Obtain pipeline inspection data and process the data format to make it meet the requirements of defect equivalent simplification. Select target defects in sequence for projection simplification. The projection simplification process is divided into axial projection and circumferential projection, and the two processes are independent of each other; (3) Determine whether the defects on the axial and circumferential projection lines overlap, and simplify the overlapping defects into equivalent single defects; based on the interaction criterion, determine whether there is interaction between the equivalent single defects on the axial and circumferential projection lines, and simplify the interaction defects into equivalent single defects that do not affect each other; (4) All the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on the neural network, and the minimum value is selected as the failure pressure at the defect. The defect failure pressure influence range is drawn based on the defect center position and the interaction distance to obtain the target pipeline failure pressure.

[0029] The failure data training set includes the failure pressure of a single defect pipeline corresponding to different pipeline and defect parameters, which is used to train the neural network. Pipeline parameters include the outer diameter and wall thickness of the pipeline, and defect parameters include the length, width and depth of the defect.

[0030] The failure pressure prediction model based on neural network has input layer nodes corresponding to pipeline outer diameter, pipeline wall thickness, defect length, defect width and defect depth, and the output is the failure pressure of a single defect pipeline.

[0031] Axial and circumferential projection of defects: According to the coordinate direction of the pipeline column, the defects are projected in the axial and circumferential directions. The projection area is divided based on the center position of the target defect and the interaction distance. The defects in the area are projected onto the axial and circumferential projection lines.

[0032] The equivalent simplification of overlapping defects and interacting defects first determines whether the defects on the axial and circumferential projection lines overlap, and simplifies the overlapping defects into equivalent single defects. Then, it determines whether there is interaction between the equivalent single defects on the axial and circumferential projection lines, and simplifies the interacting defects into equivalent single defects that do not affect each other.

[0033] The target pipeline failure pressure is composed of the failure pressure at each defect and the failure pressure influence range. The failure pressure at each defect is the output result of the prediction model, and the failure pressure influence range is the axial and circumferential influence range calculated based on the defect center position and the interaction spacing. Among them, the influence range is a rectangular range formed by the defect center position as the midpoint and the interaction spacing as the boundary.

[0034] The equivalent simplified processing process for the axial and circumferential projections of defects, as well as overlapping defects and interacting defects, includes the following steps: (1) Determine the size and location parameters of the defects on the pipeline, select the target defects in order, divide the axial and circumferential projection lines through the center of the target defect, draw the projection lines on both sides in parallel, and form a projection area together. The spacing between the axial and circumferential projection lines is Z , S : ; ; in, Z is the spacing between axial projection lines, S is the spacing between circumferential projection lines, and Represent the axial and circumferential interaction distances, respectively.

[0035] (2) If other defects are located in the axial projection area or the circumferential projection area, they are projected onto the axial projection line and the circumferential projection line respectively. If there are overlapping defects on the projection line, the overlapping defects are simplified into equivalent single defects. The length of the equivalent single defect is the total length of the defect. The depth and width of the equivalent single defect are obtained by the ratio of the effective area to the total length of the defect, as shown in the following formula: ; ; ; in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the effective length of the defect in the overlapping area, is the effective width of the defect in the overlapping area, is the effective area of ​​the defect axial section in the overlap region, is the effective area of ​​the defect radial section in the overlapping area.

[0036] (3) Determine whether there is interaction between equivalent single defects on the axial and circumferential projection lines, and simplify the interaction defects into equivalent single defects that do not affect each other, as shown in the following formula: ; ; ; in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the length of the interaction defect, is the width of the interaction defect, is the effective area of ​​the axial section of the interaction defect, is the effective area of ​​the radial section of the combined defect, is the spacing between interacting defects.

[0037] (4) All equivalent single defect parameters and corresponding pipeline parameters are input into the failure pressure prediction model based on the neural network in sequence, and the minimum value is selected as the failure pressure at the target defect.

[0038] like Figure 1 The multi-defect pipeline assessment flow chart shown is mainly divided into three parts: building a prediction model, multi-defect pipeline analysis, and target pipeline failure pressure.

[0039] S101: constructing a prediction model, including establishing finite element models of different pipelines and defect parameters, calculating the failure pressure of the finite element model of a single defect pipeline with different pipeline and defect parameters, and forming a failure data training set, training a neural network based on the failure data training set, and establishing a failure pressure prediction model based on the neural network.

[0040] S102: Multi-defect pipeline analysis, including obtaining multi-defect pipeline inspection data, processing the data format to make it meet the defect equivalent simplification requirements, selecting defects and simplifying the axial and circumferential projections of the defects, processing overlapping defects, analyzing the interaction between axial and circumferential defects, and obtaining equivalent single defect parameters that do not affect each other.

[0041] S103: All equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into a failure pressure prediction model based on a neural network to obtain a target pipeline failure pressure.

[0042] S101: Build a prediction model.

[0043] The specifications of actual pipelines vary, resulting in deviations between the prediction results and the actual results. If the prediction model is adapted to pipelines of different specifications, the prediction accuracy can be effectively improved. By establishing a single-defect pipeline finite element model with different pipelines and defect parameters, the failure pressure is calculated to form a failure data training set. The pipeline parameters include the outer diameter of the pipeline and the wall thickness of the pipeline, and the defect parameters include the length, width and depth of the defect. In addition, the finite element model requires the material parameters of the target pipeline, including yield strength and ultimate tensile strength.

[0044] A neural network is built based on computer language. The input layer nodes correspond to the pipeline outer diameter, pipeline wall thickness, defect length, defect width and defect depth respectively. The sigmoid function is used as the activation function, and the MSE is used as the loss function. The output is the failure pressure of a single-defect pipeline. The failure pressure training set is processed to enhance the robustness of the prediction model. The failure data training set is used to train the neural network, and a failure pressure prediction model based on a neural network is established.

[0045] S102: Multi-defect pipeline analysis.

[0046] The pipeline inspection data is obtained and preprocessed to meet the defect equivalent simplification requirements. The target defects are selected in sequence, the axial and circumferential projections of the defects are simplified, and the overlapping defects and interacting defects are equivalently simplified. Finally, multiple defects are simplified into equivalent single defects that do not affect each other. The specific process is as follows: like Figure 2 The multi-defect equivalent simplified flow chart is shown. The process is divided into an axial projection simplified process and a circumferential projection simplified process. The simplified processes of the two are similar and are performed simultaneously, and together constitute the analysis results. This embodiment takes the axial projection simplified process as an example for detailed description.

[0047] S201: Number the defects on the target pipeline. N The defect number i =1,2… N .

[0048] S202: Simplified axial and circumferential projections of defects: Select the target defects in sequence according to the defect number, and simplify the axial and circumferential projections of the defects. Defect projection is intended to simplify the defect distribution path, simplify multiple defects that interact in the axial and circumferential directions in multiple paths into multiple defects on the same projection line, and project the defects in the axial and circumferential directions according to the pipe column coordinate direction. In this example, only the axial projection is performed.

[0049] S203: Divide the axial projection line: Draw a projection line through the center of the target defect, draw a parallel line based on the circumferential interaction spacing in the interaction criterion, and the three projection lines form a projection area. The defects in the projection area are projected onto the axial projection line. The circumferential interaction spacing is specifically: ; in, Z represents the spacing between axial projection lines, Represents the circumferential interaction spacing.

[0050] like Figure 3 and Figure 4 The defect projection and overlapping defect processing schematic diagram shown in the figure is an elliptical defect as an example, which is used here only to describe a specific embodiment and is not intended to limit the exemplary embodiment according to the present invention.

[0051] Figure 3 shows an example of the axial projection results of a defect, Figure 4 An example of defect circumferential projection results is shown. The gray defect in the middle of the figure is the target defect. An axial projection line or a circumferential projection line is drawn through the center of the target defect. Parallel lines are drawn according to the circumferential or axial action spacing in the interaction criterion. The defects in the projection area are projected onto the axial or circumferential projection line.

[0052] S204: Defect number within the projection line j=n1,…,i,…,m1 (m1≤N) .

[0053] If the number of defects changes after projection simplification, renumber the defects.

[0054] S205: Judgment Is it greater than 1? represents the number of defects on the projection line, if If it is not greater than 1, it means that there is less than one defect in the projection line, and the failure pressure can be calculated without judging whether multiple defects have an impact. If it is greater than 1, go to the next step.

[0055] S206: Judgment Is it less than 0? Represents the distance between the defect and other defects in the projection line. If Less than 0, indicating that there is overlap between defects.

[0056] S207: Record defect number ( n , …，m ), calculate the equivalent single defect parameters: Lnm, wnm, dnm .

[0057] When defects are projected onto the axial or circumferential projection line, the defects may overlap. Simplify the overlapping defects. Confirm the overlapping defect parameters according to the defect number. Figure 3 and Figure 4 As shown in the figure, the length, depth and width of the equivalent single defect are calculated according to the effective area of ​​the overlapping defect, and the irregular overlapping defect is simplified into a regular equivalent single defect. The specific calculation process is as follows: ; ; ; in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the effective length of the defect in the overlapping area, is the effective width of the defect in the overlap region, is the effective area of ​​the defect axial section in the overlap region, is the effective area of ​​the defect radial section in the overlapping area.

[0058] S208: Renumber the defects within the projection line j=a1,…,i,…,b1 (a1≥a, b1≤b) .

[0059] Overlapping defects are simplified to equivalent single defects, the number of defects is changed, and the defect numbers are renumbered.

[0060] S209: Judgment Is it less than or equal to .

[0061] represents the spacing between equivalent single defects, represents the axial interaction distance. Less than or equal to Represents the interaction between equivalent single defects on the axial projection line.

[0062] The interaction spacing is affected by the pipe size and material parameters. The axial and circumferential interaction spacings are selected as needed. For example, in DNV-RP-F101 (a pipe standard), , .

[0063] like Figure 5 The simplified schematic diagram of the interaction defect equivalent is shown, in which the shape of the defect is taken as an ellipse as an example, which is used here only to describe a specific embodiment and is not intended to limit the exemplary embodiment according to the present invention.

[0064] The two interacting defects are a and b Distributed on the same projection line, and are the depths of two single defects, and The lengths of the two single defects, is the effective spacing between two single defects, is the length of the equivalent single defect, is the depth of the equivalent single defect. By judging whether the defects on the same projection line interact with each other, the interacting defects are simplified into equivalent single defects that do not affect each other.

[0065] S210: Record defect number ( a , …, b ), calculate the equivalent single defect parameters: Lab, wab, dab .

[0066] like Figure 5 As shown in the figure, when there is interaction between defects, the interaction defect parameters are confirmed according to the defect number, the effective area of ​​the interaction defect is calculated, and the equivalent length, depth and width of the equivalent single defect are calculated. The specific calculation process is as follows: ; ; ; in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the length of the interaction defect, is the effective area of ​​the axial section of the interaction defect, is the effective area of ​​the radial section of the combined defect, is the spacing between interacting defects.

[0067] When multiple defects interact with each other, it is necessary to arrange all combinations for simplification. After simplification, all defects are regarded as equivalent single defects that do not affect each other. Figure 5There are four defects in the figure. The three defects on the left interact with each other, while the fourth defect does not interact with other defects. According to the form of permutation and combination, the first and second defects are equivalent to single defects, the second and third defects are equivalent to single defects, and the first, second and third defects are equivalent to single defects. All three forms are regarded as simplified results of interaction defects.

[0068] S211: Equivalent single defect parameters that do not affect each other.

[0069] Through equivalent calculation, the interacting defects on the projection line are simplified into equivalent single defects that do not affect each other, and the equivalent single defect parameters are obtained.

[0070] S212: Judgment i Is it equal to N .

[0071] when i equal N When , it means that all defects have completed projection and equivalent simplification, otherwise other defects are selected for axial and circumferential projection simplification.

[0072] S213: Record all equivalent single defect parameters that do not affect each other during the calculation process.

[0073] After the equivalent calculation of overlapping defects and the equivalent calculation of interacting defects, all equivalent single defect parameters that do not affect each other obtained during the calculation process are recorded and used to predict the failure pressure of the target pipeline.

[0074] S103: Target pipeline failure pressure.

[0075] The equivalent single defect parameters and corresponding pipeline parameters that do not affect each other are input into the failure pressure prediction model based on the neural network in sequence, and the minimum value is selected as the failure pressure at the target defect. The defect failure pressure influence range is drawn based on the defect center position and the interaction distance. When the ranges overlap, the minimum failure pressure value is selected to obtain the target pipeline failure pressure.

[0076] Embodiment 2: Multi-defect pipeline failure pressure assessment system, including: The failure pressure prediction unit is configured to: calculate the failure pressure based on the single defect pipeline finite element model with different pipeline and defect parameters, form a failure data training set, train the neural network, and establish a failure pressure prediction model based on the neural network; The projection simplification unit is configured to: determine the defect parameters on the pipeline according to the inspection data of the pipeline, determine the axial projection line and the circumferential projection line through the center of the target defect respectively, and determine the projection area, and project the defects in the projection area along the axial direction and circumferential direction of the pipeline respectively; The multi-defect pipeline analysis unit is configured to: if the number of defects on the projection line exceeds one and there is overlap between the defects, then the parameters of the equivalent single defect are calculated according to the effective area of ​​the overlapping defects, and the irregular overlapping defects are simplified into regular equivalent single defects; The multi-defect pipeline analysis unit is further configured to: if there is interaction between equivalent single defects on the projection line, calculate the parameters of the equivalent single defects according to the effective area of ​​the interacting defects, and simplify the interacting defects into equivalent single defects that do not affect each other; The failure pressure prediction unit is further configured to: input all the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other into the failure pressure prediction model based on the neural network to obtain the failure pressure of the target pipeline.

[0077] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for assessing failure pressure of a multi-defect pipeline, characterized in that: The following steps are involved: Based on the finite element model of a single defect pipeline with different pipeline and defect parameters, the failure pressure is calculated to form a failure data training set. By training the neural network, a failure pressure prediction model based on the neural network is established; According to the inspection data of the pipeline, the defect parameters on the pipeline are determined, the target defects are selected in turn, the axial projection line and the circumferential projection line are determined through the center of the target defect, the projection area is determined, and the defects in the projection area are projected along the axial direction and circumferential direction of the pipeline respectively; If there is overlap between defects on the projection line, the parameters of the equivalent single defect are calculated based on the effective area of ​​the overlapping defects; If there is interaction between the equivalent single defects on the projection line, the parameters of the equivalent single defects are calculated according to the effective area of ​​the interaction defects, and the interaction defects are simplified into equivalent single defects that do not affect each other; All the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on neural network to obtain the failure pressure of the target pipeline.

2. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: A failure pressure prediction model based on a neural network is established, including: using a single-defect pipeline finite element model with different pipelines and different defect parameters to calculate the corresponding failure pressure, and the obtained failure pressure forms a failure data training set for training the neural network.

3. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: Establishing a failure pressure prediction model based on a neural network also includes: using pipeline parameters and defect parameters of a failure data training set as input data and the failure pressure of a single defective pipeline as output data to establish a failure pressure prediction model based on a neural network.

4. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: According to the inspection data of the pipeline, the defect parameters on the pipeline are determined, the target defects are selected in turn, the axial projection line and the circumferential projection line are determined through the center of the target defect, the projection area is determined, and the defects in the projection area are projected along the axial and circumferential directions of the pipeline respectively; specifically: Determine the size and location parameters of the defects on the pipeline, select the target defects in turn, determine the axial and circumferential projection lines through the center of the target defect, and form the projection area with the areas parallel to both sides of the corresponding projection lines and with a set spacing; the set spacing is the spacing between the axial and circumferential projection lines, as shown in the following formula: ； ； in, Z is the spacing between axial projection lines, S is the spacing between circumferential projection lines, and Represent the axial and circumferential interaction distances, respectively.

5. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: There is overlap between defects. Specifically, if the distance between a defect and other defects in the projection line is less than 0, there is overlap between defects.

6. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: According to the effective area of ​​overlapping defects, the parameters of equivalent single defects are calculated as shown in the following formula: ； ； ； in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the effective length of the defect in the overlapping area, is the effective width of the defect in the overlapping area, is the effective area of ​​the defect axial section in the overlap region, is the effective area of ​​the defect radial section in the overlap region, n , m Number the defect.

7. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: There is interaction between equivalent single defects, specifically: the spacing between equivalent single defects is smaller than the corresponding axial or circumferential interaction spacing.

8. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: According to the effective area of ​​the interaction defect, the parameters of the equivalent single defect are calculated as shown in the following formula: ； ； ； in, is the equivalent single defect length, is the equivalent single defect width, is the equivalent single defect depth, is the length of the interaction defect, is the width of the interaction defect, is the effective area of ​​the axial section of the interaction defect, is the effective area of ​​the radial section of the combined defect, is the spacing between interacting defects, a, b Number the defects that interact with each other.

9. The method for evaluating failure pressure of a multi-defect pipeline according to claim 1, characterized in that: All the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other are input into the failure pressure prediction model based on neural network to obtain the failure pressure of the target pipeline, which is specifically: The equivalent single defect parameters and corresponding pipeline parameters that do not affect each other are input into the failure pressure prediction model based on neural network in turn, and the minimum value is selected as the failure pressure at the target defect. The influence range of the defect failure pressure is determined based on the defect center position and the interaction distance. When the range overlaps, the minimum failure pressure value is selected to obtain the target pipeline failure pressure.

10. An evaluation system for implementing the method for evaluating failure pressure of a multi-defect pipeline according to any one of claims 1 to 9, characterized in that: include: The failure pressure prediction unit is configured to: calculate the failure pressure based on the single defect pipeline finite element model with different pipeline and defect parameters, form a failure data training set, and establish a failure pressure prediction model based on the neural network by training the neural network; The projection simplification unit is configured to: determine the defect parameters on the pipeline according to the inspection data of the pipeline, select the target defects in turn, determine the axial projection line and the circumferential projection line through the center of the target defect, determine the projection area, and project the defects in the projection area along the axial direction and circumferential direction of the pipeline respectively; The multi-defect pipeline analysis unit is configured to: if the number of defects on the projection line exceeds one and there is overlap between the defects, calculate the parameters of the equivalent single defect based on the effective area of ​​the overlapping defects; The multi-defect pipeline analysis unit is further configured to: if there is interaction between equivalent single defects on the projection line, calculate the parameters of the equivalent single defects according to the effective area of ​​the interacting defects, and simplify the interacting defects into equivalent single defects that do not affect each other; The failure pressure prediction unit is further configured to: input all the equivalent single defect parameters and corresponding pipeline parameters obtained in the calculation process that do not affect each other into the failure pressure prediction model based on the neural network to obtain the failure pressure of the target pipeline.

Images