Method, device and equipment for determining failure pressure of pipeline under double defects and medium

Abstract

The invention relates to a method, a device and equipment for determining the failure pressure of a pipeline under double defects and a medium. The method comprises the following steps: acquiring pipeline parameters and defect parameters of a to-be-processed pipeline; according to the defect parameters, the defect distance between the two defects of the to-be-processed pipeline is determined; according to the pipeline parameters, the defect parameters and the defect spacing, determining an inter-defect interaction coefficient between the two defects; when the interaction coefficient between the defects is smaller than a set value, the pipeline failure pressure of the to-be-processed pipeline is determined according to the pipeline parameters and the defect parameters; and when the interaction coefficient between the defects is not smaller than the set value, the pipeline failure pressure of the to-be-processed pipeline is determined according to the pipeline parameters. Through the method, random distribution of defects and uncertainty of shapes and sizes can be comprehensively considered, the conservative property of pipeline failure pressure evaluation under double defects is greatly reduced, the method has high reliability, and the accuracy of pipeline structure integrity evaluation can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline safety. Specifically, the present invention relates to a method, device, equipment and medium for determining the failure pressure of a pipeline under double defects. Background Art

[0002] During the long-term service of oil and gas pipelines, due to the presence of soil media and pollutants in the transported oil and gas, the pipelines are extremely vulnerable to corrosion, seriously affecting the safety of the pipelines. The continuous action of corrosion causes the pipe wall to gradually thin, and the pressure-bearing capacity of the pipeline also decreases accordingly. When the pipe wall thins to a certain extent, perforations are formed in the pipeline, causing accidents such as pipeline leakage or rupture, resulting in economic losses and safety hazards.

[0003] Corrosion is caused by the random interaction between the pipeline and surrounding chemical substances. Therefore, the position and shape of defects are uncertain and irregular. According to the number of corrosion defects, they can be divided into single-point corrosion defects and multi-point corrosion defects. At present, sufficient research has been completed on the structural integrity assessment methods for pipelines with single-point corrosion defects at home and abroad. However, for the structural integrity assessment methods of pipelines with double defects, domestic and foreign scholars mainly focus on the interaction distance between defects. For defects with interaction, the sizes and spacing of the double defects are equivalent to a single defect. The results evaluated by this method are often weaker than the actual pressure-bearing capacity of the pipeline, resulting in unnecessary pipe replacement and maintenance.

[0004] Moreover, the formation of defects is often accompanied by randomness and uncertainty, and their arrangement cannot be on the same axis or the same circumferential section of the pipeline; the currently common interaction criteria between defects are proposed based on the same defect size, and whether they are applicable for different defect sizes still needs to be studied; in addition, evaluating the double defects as a single defect undoubtedly increases the influence of the defect spacing on the pressure-bearing capacity of the pipeline. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, device, equipment and medium for determining the failure pressure of a pipeline under double defects, aiming to solve at least one of the above technical problems.

[0006] In a first aspect, the technical solution of the present invention to solve the above technical problem is as follows: A method for determining the failure pressure of a pipeline under double defects, the method comprising:

[0007] Obtain the pipeline parameters and defect parameters of the pipeline to be processed, where the pipeline to be processed contains two defects;

[0008] Determine the defect spacing between the two defects of the pipeline to be processed according to the defect parameters;

[0009] Determine the defect interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing, where the defect interaction coefficient characterizes the interaction relationship between the two defects;

[0010] When the defect interaction coefficient is less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters;

[0011] When the defect interaction coefficient is not less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters.

[0012] The beneficial effects of the present invention are as follows: Based on the defect interaction coefficient, it is determined whether there is an interaction between the two defects. Then, when determining the pipeline failure pressure of the pipeline to be processed, it can be considered whether to refer to the defect parameters. And based on this method, not only can the random distribution of defects and the uncertainty of shape and size be comprehensively considered, but also the conservativeness of the pipeline failure pressure assessment under double defects is greatly reduced, with high reliability, and the accuracy of pipeline structural integrity assessment can be improved.

[0013] On the basis of the above technical solutions, the present invention can also be improved as follows.

[0014] Further, the defect parameters include the defect length and the defect width. The determining of the defect spacing of the pipeline to be processed according to the defect parameters includes:

[0015] Take the defect with the shortest axial distance from the pipeline to be processed to the girth weld of the pipeline to be processed as the first defect, and determine the first defect center of this first defect;

[0016] Establish a local coordinate system with the first defect center as the origin;

[0017] Determine the distance between each defect on the pipeline to be processed and the origin;

[0018] According to the defect length, the defect width and each of the distances, determine the axial defect spacing and the circumferential defect spacing between the defects. The defect spacing includes the axial defect spacing and the circumferential defect spacing between the defects.

[0019] Further, the pipeline parameters include the pipeline diameter, wall thickness, ultimate tensile strength and hardening index of the pipeline material, the defect parameters include the defect length, defect width and defect depth, and the defect spacing includes the axial defect spacing and the circumferential defect spacing between the defects; the determining of the defect interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing includes:

[0020] Taking any one of the pipeline parameters, any one of the defect parameters, and any one of the defect spacings as a parameter combination to obtain multiple groups of parameter combinations;

[0021] For any group of parameter combinations, determining the interaction coefficient between the two defects under the action of this parameter combination;

[0022] Determining the interaction coefficient between the two defects according to the interaction coefficients between the defects corresponding to each of the parameter combinations.

[0023] Further, the determining the interaction coefficient between the two defects according to the pipeline parameters, the defect parameters, and the defect spacing includes:

[0024] Determining the interaction coefficient between the two defects according to the pipeline parameters, the defect parameters, and the defect spacing through a pre-trained prediction model, and the prediction model is trained based on an artificial neural network.

[0025] Further, obtaining training samples, where the training samples include pipeline parameters, defect parameters, and the defect spacing between any two defects of a pipeline containing multiple defects. The pipeline parameters include pipeline diameter, wall thickness, ultimate tensile strength, and hardening index of the pipeline material. Each defect parameter includes defect length, defect width, and defect depth. The defect spacing includes the axial spacing and circumferential spacing between the defects;

[0026] Taking any one of the pipeline parameters, any one of the defect parameters, and any one of the defect spacings in the training samples as a parameter combination to obtain multiple groups of sample parameter combinations;

[0027] For any two defects, based on the multiple groups of sample parameter combinations corresponding to the two defects, simulating through a finite element model to obtain the true interaction coefficient between the two defects under each group of the sample parameter combinations;

[0028] For any two defects, based on the multiple groups of sample parameter combinations corresponding to the two defects, obtaining the predicted interaction coefficient between the two defects under each group of the sample parameter combinations through an artificial neural network;

[0029] Training the prediction model according to each of the true interaction coefficients between the defects and each of the predicted interaction coefficients between the defects.

[0030] Further, before obtaining the predicted interaction coefficient between the two defects under each group of the sample parameter combinations through an artificial neural network based on the multiple groups of sample parameter combinations corresponding to the two defects, the method further includes:

[0031] For any two defects, Gaussian white noise is added to multiple sets of sample parameter combinations corresponding to the two defects to obtain data to be trained.

[0032] In a second aspect, the present invention also provides a device for determining the failure pressure of a pipeline under double defects to solve the above technical problems. The device includes:

[0033] An acquisition module, configured to acquire pipeline parameters and defect parameters of a pipeline to be processed, where the pipeline to be processed includes two defects;

[0034] A defect spacing determination module, configured to determine the defect spacing between the two defects of the pipeline to be processed according to the defect parameters;

[0035] A coefficient determination module, configured to determine the interaction coefficient between defects between the two defects according to the pipeline parameters, the defect parameters, and the defect spacing, where the interaction coefficient between defects characterizes the interaction relationship between the two defects;

[0036] A pipeline failure pressure determination module, configured to determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters when the interaction coefficient between defects is less than a set value; and determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters when the interaction coefficient between defects is not less than the set value.

[0037] In a third aspect, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method for determining the failure pressure of a pipeline under double defects of the present application is implemented.

[0038] In a fourth aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for determining the failure pressure of a pipeline under double defects of the present application is implemented.

[0039] Additional aspects and advantages of the present application will be given in part in the following description, and these will become apparent from the following description, or can be understood through the practice of the present application. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments of the present invention will be briefly introduced below.

[0041] Figure 1 It is a schematic flowchart of a method for determining the failure pressure of a pipeline under double defects provided by an embodiment of the present invention;

[0042] Figure 2 Geometric schematic diagram of a double defect with staggered pipelines provided in an embodiment of the present invention;

[0043] Figure 3 Schematic diagram of a prediction curve of an artificial neural network provided in an embodiment of the present invention;

[0044] Figure 4 Schematic diagram of an error curve of an artificial neural network provided in an embodiment of the present invention;

[0045] Figure 5 Schematic diagram of the structure of a pipeline failure pressure determination device under double defects provided in an embodiment of the present invention;

[0046] Figure 6 Schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed implementation manners

[0047] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not intended to limit the scope of the present invention.

[0048] The technical solutions of the present invention and how the technical solutions of the present invention solve the above technical problems are described in detail below with specific embodiments. These several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below with reference to the accompanying drawings.

[0049] The solutions provided in the embodiments of the present invention can be applied to any application scenario that needs to determine the failure pressure of a pipeline containing double defects. The solutions provided in the embodiments of the present invention can be executed by any electronic device, including at least one of the following: smart phones, tablet computers, laptop computers, desktop computers, smart speakers, smart watches, smart TVs, and smart vehicle-mounted devices.

[0050] An embodiment of the present invention provides a possible implementation manner, as Figure 1 shown, a flowchart of a method for determining the failure pressure of a pipeline under double defects is provided. This solution can be executed by any electronic device. For example, it can be a terminal device, or jointly executed by a terminal device and a server. For the convenience of description, the method provided in the embodiments of the present invention will be described below with a terminal device as the execution subject. As Figure 1 shown in the flowchart, the method may include the following steps:

[0051] Step S10, obtaining the pipeline parameters and defect parameters of the pipeline to be processed, where the pipeline to be processed contains two defects;

[0052] Step S20: Determine the defect spacing between two defects of the pipeline to be processed according to the defect parameters.

[0053] Step S30: Determine the interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing. The interaction coefficient between the defects characterizes the interaction relationship between the two defects.

[0054] Step S40: When the interaction coefficient between the defects is less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters.

[0055] Step S50: When the interaction coefficient between the defects is not less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters.

[0056] Through the method of the present invention, it is determined whether there is an interaction between two defects based on the interaction coefficient between the defects. Then, when determining the pipeline failure pressure of the pipeline to be processed, it is possible to consider whether to refer to the defect parameters. And based on this method, not only can the random distribution of defects and the uncertainty of shape and size be comprehensively considered, but also the conservativeness of the pipeline failure pressure assessment under double defects is greatly reduced, with high reliability, and the accuracy of pipeline structural integrity assessment can be improved.

[0057] The following further illustrates the solution of the present invention in combination with the following specific embodiments. In this embodiment, a method for determining the pipeline failure pressure under double defects may include the following steps:

[0058] Step S10: Obtain the pipeline parameters and defect parameters of the pipeline to be processed. The pipeline to be processed contains two defects, which can be double defects arranged staggeredly.

[0059] Among them, the pipeline parameters may include but are not limited to the pipeline diameter, wall thickness, ultimate tensile strength and hardening index of the pipeline material. The defect parameters include defect length, defect width and defect depth. For the two defects involved in this solution, the defect parameters of each defect may be different or the same.

[0060] See Figure 2 the geometric schematic diagram of the double defects with staggered arrangement of the pipeline shown in Figure 2 where L in p represents the pipeline length, Figure 2 where t in Figure 2 represents the wall thickness, and each of the three dashed boxes shown in c includes two defects. S L represents the circumferential defect spacing, and S

[0061] For a pipeline containing more than two defects, for any two of the more than two defects, the method of this solution can also be used to determine the pipeline failure pressure of the pipeline to be processed.

[0062] Step S20: Determine the defect spacing between the two defects of the pipeline to be processed according to the defect parameters.

[0063] Optionally, one implementation of the above S20 is as follows:

[0064] S201: Take the defect with the shortest axial distance from the circumferential weld of the pipeline to be processed as the first defect, and determine the center of the first defect.

[0065] S202: Establish a local coordinate system with the center of the first defect as the origin.

[0066] S203: Determine the distance between each defect on the pipeline to be processed and the origin.

[0067] S204: Determine the axial spacing and circumferential spacing between defects according to the defect length, the defect width and each of the distances. The defect spacing includes the axial spacing and circumferential spacing between defects.

[0068] Step S30: Determine the interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing. The interaction coefficient between defects characterizes the interaction relationship between the two defects. The larger the interaction coefficient between defects, the more interaction exists between the two defects, and the defect parameters will affect the determination of the pipeline failure pressure of the pipeline to be processed. On the contrary, the smaller the interaction coefficient between defects, the less interaction exists between the two defects, which can be ignored, and the defect parameters will not affect the determination of the pipeline failure pressure of the pipeline to be processed. Then, the pipeline failure pressure of the pipeline to be processed can be determined based on the following two steps.

[0069] Step S40: When the interaction coefficient between defects is less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters.

[0070] Step S50: When the interaction coefficient between defects is not less than the set value, determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters.

[0071] Among them, determining the pipeline failure pressure of the pipeline to be processed based on the pipeline parameters and the defect parameters, and determining the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters can both be realized based on the prior art, which will not be elaborated here.

[0072] Optionally, one implementation of S30 is as follows:

[0073] S301, take any one parameter from the pipeline parameters, any one parameter from the defect parameters, and any one parameter from the defect spacing as a parameter combination to obtain multiple parameter combinations;

[0074] S302, for any parameter combination, determine the interaction coefficient between the two defects under the action of this parameter combination;

[0075] S303, according to the interaction coefficients between defects corresponding to each parameter combination (for example, weighted fusion), determine the interaction coefficient between the two defects.

[0076] Optionally, another implementation of S30 can also be:

[0077] According to the pipeline parameters, the defect parameters, and the defect spacing, determine the interaction coefficient between the two defects through a pre-trained prediction model, and the prediction model is trained based on an artificial neural network.

[0078] Optionally, the above prediction model is trained based on the following method:

[0079] S1, obtain training samples, where the training samples include the pipeline parameters, defect parameters, and the defect spacing between any two defects of a pipeline containing multiple defects. The pipeline parameters include the pipeline diameter, wall thickness, ultimate tensile strength, and hardening index of the pipeline material. Each defect parameter includes the defect length, defect width, and defect depth. The defect spacing includes the axial spacing and circumferential spacing between the defects;

[0080] As an example, the pipeline in the training sample has a pipeline diameter of 357 mm, a pipeline wall thickness of 10 mm, and a steel grade of X52.

[0081] S2, take any one parameter from the pipeline parameters, any one parameter from the defect parameters, and any one parameter from the defect spacing in the training sample as a parameter combination to obtain multiple sample parameter combinations;

[0082] Optionally, a full-diameter model with staggered double defects can be constructed based on the finite element numerical simulation method to obtain 2000 sets of simulation data (sample parameter combinations). As an example, the characteristic variables in each group of sample parameter combinations include: defect length (3 types), defect depth (7 types), axial spacing between defects (10 types), and circumferential spacing (10 types). 2000 sets of sample parameter combinations can be obtained through data combination. For details, please refer to Table 1.

[0083] Table 1 Defect example parameters

[0084]

[0085] S3. For any two defects, based on multiple groups of sample parameter combinations corresponding to the two defects, the true interaction coefficient between the two defects under each group of the sample parameter combinations is obtained through finite element model simulation;

[0086] Optionally, different sample parameter combinations may correspond to different finite element models. Then, for any two defects, based on each group of sample parameter combinations, through their respective corresponding finite element models, the interaction coefficient between the defects corresponding to each group of sample parameter combinations can be obtained. For details, see Table 2.

[0087] Table 2 Calculation of Correlation Coefficient between Defects for Different Sample Parameter Combinations (Partial Results)

[0088]

[0089] The interaction coefficient between the defects can be the ratio of the pipe burst pressure under double defects to the pipe burst pressure under a single defect. Then, based on this, the interaction coefficient between the defects corresponding to each group of sample parameter combinations can also be determined. Then, based on the finite element numerical simulation method, each group of sample parameter combinations and the interaction coefficient between the defects corresponding to each group of sample parameter combinations are simulated and trained to obtain a full-diameter finite element model (also known as a finite element model) with staggered defects under different defect sizes and pipe specifications (different sample parameter combinations), forming a corresponding data set.

[0090] When the interaction coefficient between the defects is less than a set value, for example, less than 0.99, there is an interaction between the defects. While when the interaction coefficient between the defects exceeds the set value, for example, exceeds 0.99, it is considered that the interaction between the defects can be ignored. Correspondingly, its influence on the pipe failure pressure is also ignored.

[0091] S4. For any two defects, based on multiple groups of sample parameter combinations corresponding to the two defects, the predicted interaction coefficient between the two defects under each group of the sample parameter combinations is obtained through an artificial neural network;

[0092] S5. The prediction model is trained according to each of the true interaction coefficients between the defects and each of the predicted interaction coefficients between the defects.

[0093] Essentially, each group of sample parameter combinations and the corresponding true interaction coefficients between the defects can be used as a training set to train the artificial neural network to obtain the predicted interaction coefficient between the two defects under each group of the sample parameter combinations.

[0094] As an example, 2000 groups of data set results (interaction coefficients between sample parameter combinations and corresponding real defects) are selected as the training set, and the artificial neural network algorithm is used to predict the interaction coefficients between staggered double defects under different pipe materials, defect sizes and defect spacings. When using the artificial neural network (Artificial Neural Network, i.e., ANN) model algorithm to predict the interaction coefficients between double defects, parameter settings are required: the initial learning rate is set to 0.003; the first moment variable is set to 0.9; the second moment variable is set to 0.999; the mini-batch is set to 128; the number of iterations is set to 1000; the acceptance convergence region is set to [0.0001, 0.001]. The comparison results between the predicted results and the real results can be seen in Figure 3 and Figure 4 .

[0095] Optionally, after obtaining multiple groups of sample parameter combinations, these data can be preprocessed first. One implementation method of preprocessing is:

[0096] For any two defects, Gaussian white noise is added to the multiple groups of sample parameter combinations corresponding to the two defects to obtain the data to be trained, so as to improve the robustness of the artificial neural network.

[0097] After preprocessing, the data to be trained is divided into a training set and a validation set according to a ratio.

[0098] As an example, assume that the amount of data to be trained is D, and it is randomly divided into a training set Dtrain and a validation set Deval according to a ratio of 9:1. Specifically:

[0099] D = {(x 1 , y 1 ), (x 2 , y 2 ), …, (x N , y N )}

[0100] D train = {(x 1 , y 1 ), (x 2 , y 2 ), …, (x T , y T )} ∈ D

[0101] D eval = {(x 1 , y 1 ), (x 2 , y 2 ), …, (x E , y E )} ∈ D

[0102] D train ∪D eval =D

[0103]

[0104] N = T + E

[0105] T:E = 9:1

[0106] In the above, for any two defects, adding Gaussian white noise to the multiple sets of sample parameter combinations corresponding to the two defects to obtain the specific implementation process of the data to be trained can be as follows:

[0107] For any two defects, adding Gaussian white noise ε i to the feature F of the multiple sets of sample parameter combinations corresponding to the two defects i to obtain the data to be trained D' train , thereby increasing the robustness of the artificial neural network model.

[0108] D′ train ={(x′ 1 , y 1 ), (x′ 2 , y 2 ), …, (x′ T , y T )}

[0109] ={(x 1 + ε 1 , y 1 ), (x 2 + ε 2 , y 2 ), …, (x T + ε T , y T )}

[0110] In the solution of this application, considering the discrete variable characteristics of pipeline materials, a word embedding model can be constructed through the Python language to convert the feature variables into low-dimensional continuous vectors, which are used as one of the features input to the interaction coefficient prediction model for staggered defects.

[0111] Among them, the input of the word embedding model is the one-hot encoding of pipeline materials (such as X52, X65, etc.), as shown in Table 3:

[0112] Table 3 One-hot encoding format of different pipeline materials

[0113]

[0114] The output of the word embedding model is a vector with a dimension of Nv The vectors, where N v = 8, specifically:

[0115] X o ' = ωX o

[0116] where X o ' is the output vector of the word embedding model; ω is the one-hot encoding conversion matrix, with dimensions [N v , N v ; X o is the one-hot encoding of any pipeline material.

[0117] Build an Artificial Neural Network (ANN) model algorithm based on the Python language, which can also be called a polynomial neural network model algorithm, to obtain the model framework. This model uses the pipe outer diameter, wall thickness, material, defect size, and defect spacing as feature variables to predict the interaction coefficient of staggered defects, that is, to predict the interaction coefficient between defects. Construct a loss function and a loss function optimization method, train the model based on the data to be trained, and judge the convergence effect of the model based on the validation set data to obtain better model parameters.

[0118] Among them, the process of predicting the interaction coefficient of staggered defects is specifically as follows:

[0119]

[0120]

[0121] Among them, the input vector X' is a combination of 0th-degree polynomials, 1st-degree polynomials,..., nth-degree polynomials of features such as pipe outer diameter, wall thickness, material, and defect size, where n = 3; the input layer of the polynomial neural network contains 680 neurons; the output layer contains 1 neuron; there are 3 hidden layers in total, and the number of neurons in each layer is 16, 32, and 16 respectively. The input layer and the hidden layer use the Rectified Linear Unit (ReLU) as the activation function f ha for each layer; the output layer uses the f opta function as the activation function. w 1 , w 2 , w 3 and w 4 are weights, and b 1 , b 2 , b 3 and b 4 are bias terms.

[0122] The input layer of the artificial neural network contains 7 neurons; the output layer contains 1 neuron; there are 3 hidden layers, and the number of neurons in each layer is 16, 32, and 16 respectively. The input layer and the hidden layers use the Rectified Linear Unit (ReLU) as the activation function for each layer; the output layer uses the Sigmoid function as the activation function. The specific algorithm model is as follows:

[0123] out 1 = ReLU(w 1 x + b 1 )

[0124] out 2 = ReLU(w 2 out 1 + b 2 )

[0125] out 3 = ReLU(w 3 out 2 + b 3 )

[0126]

[0127] The mean squared error function (MSE) is constructed using the output result of the artificial neural network (predicting the interaction coefficient of the staggered arrangement defects) and the simulated data label value (the interaction coefficient of the actual staggered arrangement defects) as the loss function.

[0128] The Adam (Adaptive Moment Estimation) algorithm is selected, and the data to be trained D' train is divided into small batches of data to iteratively update the weights of the artificial neural network; the change trend of the loss function result calculated by the validation set Deval with the number of iterations is used to judge the convergence trend of the artificial neural network, and whether the finally stabilized convergence value falls within the acceptance region is used to judge the overall convergence effect of the model. The specific optimization process of the Adam optimizer is as follows:

[0129] g t = ▽ θ J(θ t

[0130] m t = β 1 m t-1 + (1 - β 1 )g t

[0131] v t = β2 v t-1 +(1 - β 2 )g t 2

[0132]

[0133]

[0134] where β 1 is the first - order moment variable, i.e., the decay rate of the first - order matrix, and β 1 can be set to 0.9; β 2 is the second - order moment variable, set to 0.999; t is the time step; m t is the updated biased first - order moment estimate, i.e., the first - order moment estimate corresponding to time t; m t-1 is the first - order moment estimate at t - 1;

[0135] v t is the updated biased second - order moment estimate, i.e., the second - order moment estimate corresponding to time t; v t-1 is the second - order moment estimate at t - 1; J(θ t ) is the loss function, where θ t is the loss function parameter, ▽ θ J(θ t ) is the gradient of the loss function with respect to this parameter; θ t+1 is the updated loss function parameter. is the corrected first - order moment estimate value; is the corrected second - order moment estimate.

[0136] As an example, the initial learning rate α is set in the range of [0.001, 0.005]; ò is the algorithm optimization constant term, set to 10 -8 ; the mini - batch for neural network training is set to 128, and the last batch that does not meet the data volume of 128 is discarded and enters the next iteration.

[0137] Specific implementation plan: The loss value obtained based on the validation set is used as the evaluation index of the model. The closer the loss value of the validation set is to 0, the smaller the difference between the prediction result (the interaction coefficient between predicted defects) and the actual result (the interaction coefficient between real defects), indicating good training effect. The evaluation result of the validation set of the prediction model established by this plan is 0.000412, indicating a high fitting degree of the interaction coefficient between defects, a small error between the predicted value and the actual value, and strong reliability of the model. It shows that the neural network for predicting the interaction coefficient of axial - arranged double defects in the pipeline is completed.

[0138] The interaction coefficient between defects obtained by using the prediction model in this plan can be seen in Table 4.

[0139] Table 4 Interaction coefficients between defects with different defect parameters (partial results)

[0140]

[0141] Through the solution of the present invention, compared with the prior art, the following advantages are achieved:

[0142] Aiming at the problem that the structural integrity assessment method for pipelines with double defects cannot accurately evaluate the pressure-bearing capacity of pipelines, a prediction method for the interaction coefficient between staggered double defects in pipelines based on artificial neural network is proposed, which mainly includes four links: collection of pipeline and defect parameter information, construction of finite element model, training data based on artificial neural network model, and algorithm verification. This patent constructs a finite element model of a pipeline with staggered double defects by means of finite element software to obtain multiple groups of simulation data results under different characteristic variables; in addition, based on machine learning algorithms, an algorithm framework is constructed, the simulation data is trained, and the effectiveness of the algorithm framework is verified by means of test data.

[0143] The advantage of the present invention lies in providing a prediction method for the interaction coefficient between staggered double defects in pipelines based on artificial neural network. This method can not only comprehensively consider the random distribution of defects and the uncertainty of shape and size, but also the predicted results greatly reduce the conservatism of the pipeline failure pressure assessment under double defects, have high reliability, and can improve the accuracy of pipeline structural integrity assessment.

[0144] Based on the same principle as the method shown in Figure 1 , an embodiment of the present invention also provides a device 20 for determining the pipeline failure pressure under double defects, as shown in Figure 5 . The device 20 for determining the pipeline failure pressure under double defects may include an acquisition module 210, a defect spacing determination module 220, a coefficient determination module 230, and a pipeline failure pressure determination module 240, wherein:

[0145] The acquisition module 210 is configured to acquire pipeline parameters and defect parameters of a pipeline to be processed, and the pipeline to be processed includes two defects;

[0146] The defect spacing determination module 220 is configured to determine the defect spacing between the two defects of the pipeline to be processed according to the defect parameters;

[0147] The coefficient determination module 230 is configured to determine the interaction coefficient between the two defects according to the pipeline parameters, the defect parameters, and the defect spacing, and the interaction coefficient between the defects characterizes the interaction relationship between the two defects;

[0148] The pipeline failure pressure determination module 240 is configured to determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters when the defect interaction coefficient is less than a set value; and determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters when the defect interaction coefficient is not less than the set value.

[0149] Optionally, the above defect parameters include the defect length and the defect width. The defect spacing determination module 220 is specifically configured to:

[0150] Take the defect with the shortest axial distance from the circumferential weld of the pipeline to be processed as the first defect, and determine the first defect center of the first defect;

[0151] Establish a local coordinate system with the first defect center as the origin;

[0152] Determine the distance between each defect on the pipeline to be processed and the origin;

[0153] Determine the axial spacing and circumferential spacing between defects according to the defect length, the defect width and each of the distances. The defect spacing includes the axial spacing and circumferential spacing between defects.

[0154] Optionally, the above pipeline parameters include the pipeline diameter, wall thickness, ultimate tensile strength and hardening index of the pipeline material, the defect parameters include the defect length, defect width and defect depth, and the defect spacing includes the axial spacing and circumferential spacing between defects; when the coefficient determination module 230 determines the defect interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing, it is specifically configured to:

[0155] Take any one of the pipeline parameters, any one of the defect parameters, and any one of the defect spacing as a parameter combination to obtain multiple parameter combinations;

[0156] For any one of the parameter combinations, determine the defect interaction coefficient between the two defects under the action of the parameter combination;

[0157] Determine the defect interaction coefficient between the two defects according to the defect interaction coefficients corresponding to each of the parameter combinations.

[0158] Optionally, when the coefficient determination module 230 determines the defect interaction coefficient between the two defects according to the pipeline parameters, the defect parameters and the defect spacing, it is specifically configured to:

[0159] Determine the defect - to - defect interaction coefficient between the two defects according to the pipeline parameters, the defect parameters, and the defect spacing, where the prediction model is obtained by training based on an artificial neural network.

[0160] Optionally, the above - mentioned prediction model is obtained by training in the following manner:

[0161] Obtain training samples, where the training samples include the pipeline parameters, defect parameters, and the defect spacing between any two defects of a pipeline containing multiple defects. The pipeline parameters include the pipeline diameter, wall thickness, ultimate tensile strength, and hardening index of the pipeline material. Each defect parameter includes the defect length, defect width, and defect depth. The defect spacing includes the axial spacing and circumferential spacing between the defects;

[0162] Take any one parameter from the pipeline parameters, any one parameter from the defect parameters, and any one parameter from the defect spacing in the training samples as a parameter combination to obtain multiple groups of sample parameter combinations;

[0163] For any two defects, based on the multiple groups of sample parameter combinations corresponding to the two defects, simulate the true defect - to - defect interaction coefficient between the two defects under each group of sample parameter combinations through a finite - element model;

[0164] For any two defects, based on the multiple groups of sample parameter combinations corresponding to the two defects, obtain the predicted defect - to - defect interaction coefficient between the two defects under each group of sample parameter combinations through an artificial neural network;

[0165] Train the prediction model according to each of the true defect - to - defect interaction coefficients and each of the predicted defect - to - defect interaction coefficients.

[0166] Optionally, before obtaining the predicted defect - to - defect interaction coefficient between the two defects under each group of sample parameter combinations through an artificial neural network based on the multiple groups of sample parameter combinations corresponding to the two defects, the device further includes:

[0167] A pre - processing module, which is used to add Gaussian white noise to the multiple groups of sample parameter combinations corresponding to any two defects to obtain the data to be trained.

[0168] The pipeline failure pressure determination device under double defects according to the embodiments of the present invention can execute the pipeline failure pressure determination method under double defects provided by the embodiments of the present invention, and their implementation principles are similar. The actions performed by each module and unit in the pipeline failure pressure determination device under double defects in each embodiment of the present invention correspond to the steps in the pipeline failure pressure determination method under double defects in each embodiment of the present invention. For the detailed function descriptions of each module of the pipeline failure pressure determination device under double defects, reference can specifically be made to the descriptions in the corresponding pipeline failure pressure determination method shown above, which will not be elaborated here.

[0169] Among them, the above-mentioned pipeline failure pressure determination device under double defects can be a computer program (including program code) running in a computer device. For example, the pipeline failure pressure determination device under double defects is an application software; this device can be used to execute the corresponding steps in the method provided by the embodiments of the present invention.

[0170] In some embodiments, the pipeline failure pressure determination device under double defects provided by the embodiments of the present invention can be implemented in a combination of software and hardware. As an example, the pipeline failure pressure determination device under double defects provided by the embodiments of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the pipeline failure pressure determination method provided by the embodiments of the present invention. For example, a processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs, Application Specific Integrated Circuit), DSPs, programmable logic devices (PLDs, Programmable Logic Device), complex programmable logic devices (CPLDs, Complex ProgrammableLogic Device), field-programmable gate arrays (FPGAs, Field-Programmable Gate Array) or other electronic components.

[0171] In other embodiments, the pipeline failure pressure determination device under double defects provided by the embodiments of the present invention can be implemented in a software manner. Figure 5 Shows the pipeline failure pressure determination device stored in the memory, which can be software in the form of programs and plugins, etc., and includes a series of modules, including an acquisition module 210, a defect spacing determination module 220, a coefficient determination module 230, and a pipeline failure pressure determination module 240, for implementing the pipeline failure pressure determination method provided by the embodiments of the present invention.

[0172] The modules involved in the embodiments of the present invention can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the module itself in some cases.

[0173] Based on the same principle as the method shown in the embodiments of the present invention, an electronic device is also provided in the embodiments of the present invention. The electronic device may include, but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the method shown in any embodiment of the present invention by calling the computer program.

[0174] In an alternative embodiment, an electronic device is provided, such as Figure 6 shown Figure 6 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. Among them, the processor 4001 and the memory 4003 are connected, such as connected through a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, and the transceiver 4004 may be used for data interaction between the electronic device and other electronic devices, such as data sending and / or data receiving, etc. It should be noted that in practical applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation to the embodiments of the present invention.

[0175] The processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in connection with the disclosure of the present invention. The processor 4001 may also be a combination for implementing computing functions, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0176] The bus 4002 may include a path for transmitting information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation,Figure 6 It is represented only by a thick line, but it does not mean that there is only one bus or one type of bus.

[0177] The memory 4003 can be a ROM (Read Only Memory), or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory), or other types of dynamic storage devices that can store information and instructions. It can also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0178] The memory 4003 is used to store the application program code (computer program) for implementing the solution of the present invention and is controlled by the processor 4001 for execution. The processor 4001 is used to execute the application program code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0179] Among them, the electronic device can also be a terminal device. Figure 6 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

[0180] The embodiments of the present invention provide a computer-readable storage medium, on which a computer program is stored. When the computer program runs on a computer, it enables the computer to execute the corresponding content in the foregoing method embodiments.

[0181] According to another aspect of the present invention, there is also provided a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the above various implementation manners.

[0182] Computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0183] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0184] The computer-readable storage medium provided by the embodiments of the present invention may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0185] The above computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to execute the method shown in the above embodiments.

[0186] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present invention.

Claims

1. A method for determining failure pressure of a pipeline under double defects, characterized in that: The following steps are involved: Obtaining pipeline parameters and defect parameters of a pipeline to be processed, wherein the pipeline to be processed includes two defects; Determining the defect distance between two defects of the pipeline to be processed according to the defect parameters; Determining an inter-defect interaction coefficient between the two defects according to the pipeline parameter, the defect parameter and the defect distance, wherein the inter-defect interaction coefficient represents an interaction relationship between the two defects; When the interaction coefficient between defects is less than a set value, determining the pipeline failure pressure of the pipeline to be treated according to the pipeline parameters and the defect parameters; When the interaction coefficient between defects is not less than the set value, the pipeline failure pressure of the pipeline to be processed is determined according to the pipeline parameters.

2. The method according to claim 1, characterized in that The defect parameters include defect length and defect width, and determining the defect spacing of the pipeline to be processed according to the defect parameters includes: Taking the defect of the pipeline to be processed that is closest to the girth weld of the pipeline to be processed in the axial direction as the first defect, and determining the first defect center of the first defect; Establishing a local coordinate system with the first defect center as the origin; Determining the distance between each defect on the pipeline to be processed and the origin; According to the defect length, the defect width and each of the distances, an axial spacing between defects and an annular spacing between defects are determined, and the defect spacing includes the axial spacing between defects and the annular spacing between defects.

3. The method according to claim 1, characterized in that The pipeline parameters include pipeline diameter, wall thickness, ultimate tensile strength and hardening index of pipeline material, the defect parameters include defect length, defect width and defect depth, and the defect spacing includes axial spacing and circumferential spacing between defects; The step of determining the defect interaction coefficient between the two defects according to the pipeline parameter, the defect parameter and the defect distance comprises: Taking any one of the pipeline parameters, any one of the defect parameters, and any one of the defect spacing parameters as a parameter combination to obtain multiple groups of parameter combinations; For any set of parameter combinations, determining the defect interaction coefficient between the two defects under the effect of the parameter combination; The defect interaction coefficient between the two defects is determined according to the defect interaction coefficient corresponding to each parameter combination.

4. The method according to any one of claims 1 to 3, characterized in that The step of determining the defect interaction coefficient between the two defects according to the pipeline parameter, the defect parameter and the defect distance comprises: According to the pipeline parameters, the defect parameters and the defect spacing, the defect interaction coefficient between the two defects is determined by a pre-trained prediction model, wherein the prediction model is obtained based on artificial neural network training.

5. The method according to claim 4, characterized in that The prediction model is trained based on the following method: Acquire training samples, wherein the training samples include pipeline parameters including multiple defects, defect parameters, and defect spacing between any two defects, wherein the pipeline parameters include pipeline diameter, wall thickness, ultimate tensile strength and hardening index of pipeline materials, each defect parameter includes defect length, defect width and defect depth, and the defect spacing includes axial spacing and circumferential spacing between defects; Taking any one of the pipeline parameters in the training sample, any one of the defect parameters, and any one of the defect spacing parameters as a parameter combination to obtain multiple groups of sample parameter combinations; For any two defects, based on multiple groups of sample parameter combinations corresponding to the any two defects, a true defect interaction coefficient between the any two defects under each group of sample parameter combinations is obtained through finite element model simulation; For any two defects, based on multiple groups of sample parameter combinations corresponding to the any two defects, a predicted defect interaction coefficient between the any two defects under each group of sample parameter combinations is obtained through an artificial neural network; The prediction model is obtained by training according to the interaction coefficients between the real defects and the interaction coefficients between the predicted defects.

6. The method according to claim 5, characterized in that Before obtaining the predicted defect interaction coefficient between any two defects under each set of sample parameter combinations based on the multiple sets of sample parameter combinations corresponding to the any two defects through an artificial neural network, the method further includes: For any two defects, Gaussian white noise is added to multiple groups of sample parameter combinations corresponding to the any two defects to obtain data to be trained.

7. A device for determining failure pressure of a pipeline under double defects, characterized in that: include: An acquisition module, used to acquire pipeline parameters and defect parameters of a pipeline to be processed, wherein the pipeline to be processed includes two defects; A defect spacing determination module, used to determine the defect spacing between two defects of the pipeline to be processed according to the defect parameters; A coefficient determination module, used to determine the defect interaction coefficient between the two defects according to the pipeline parameter, the defect parameter and the defect spacing, wherein the defect interaction coefficient represents the interaction relationship between the two defects; The pipeline failure pressure determination module is used to determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters and the defect parameters when the interaction coefficient between defects is less than a set value; and to determine the pipeline failure pressure of the pipeline to be processed according to the pipeline parameters when the interaction coefficient between defects is not less than the set value.

8. The device according to claim 7, characterized in that The defect parameters include defect length and defect width, and the defect spacing determination module is specifically used to: Taking the defect of the pipeline to be processed that is closest to the girth weld of the pipeline to be processed in the axial direction as the first defect, and determining the first defect center of the first defect; Establishing a local coordinate system with the first defect center as the origin; Determining the distance between each defect on the pipeline to be processed and the origin; According to the defect length, the defect width and each of the distances, an axial spacing between defects and an annular spacing between defects are determined, and the defect spacing includes the axial spacing between defects and the annular spacing between defects.

9. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images