CN113378420A

CN113378420A - Method for predicting service life of crack pipeline

Info

Publication number: CN113378420A
Application number: CN202011359750.XA
Authority: CN
Inventors: 谢明江; 苏春; 张辉; 邢佑强; 吴泽
Original assignee: Southeast University
Current assignee: Southeast University
Priority date: 2020-11-27
Filing date: 2020-11-27
Publication date: 2021-09-10

Abstract

The invention belongs to the technical field of pipeline service life prediction, and particularly relates to an integrated prediction method for a crack pipeline. The shape of the pipeline crack is semi-elliptical, the fatigue crack grows along with the variable load in the pipeline, and the initial size of the crack is detected by an online detection tool. The method integrates the characteristics of a traditional Paris law crack growth model and online detection data, and improves the accuracy of the prediction model by updating the parameters of the prediction physical model, so that more accurate service life prediction is carried out. The service life prediction method can be widely applied to operation and maintenance management of the crack pipeline.

Description

Method for predicting service life of crack pipeline

Technical Field

The invention belongs to the technical field of pipeline service life prediction, and relates to an integrated prediction method for a crack pipeline.

Background

Pipelines are considered the safest and most economical way to transport large quantities of oil and gas products. 94% of the refined petroleum products and most of the oil and gas outlets are transported by pipeline. Pipes present different types of defects such as fatigue cracks, corrosion, etc. Without proper remedial action, these defects can ultimately lead to pipe failure, including leaks or ruptures, which can result in very expensive down time, as well as leakage and leakage of the pipe contents into the environment. Currently, pipeline companies commonly perform on-line inspections periodically using on-line inspection tools to detect defects and assess pipeline health. Fatigue cracks refer to the propagation of fatigue cracks during pipe operation due to pressure cycling.

Fatigue cracks are a main defect type of pipelines, the existing prediction model of the service life of the cracked pipelines has larger errors, and the treatment of the fatigue cracks is always an important task for managing the integrity of the pipelines. However, existing in-line inspection (ILI) tools have significant uncertainty in fatigue crack measurements and typically have an error of about plus or minus 1 mm, with 80% confidence. In addition, the current crack propagation model method based on the Paris law is mainly used for fatigue crack propagation life prediction. Uncertainty in crack size and life prediction physical models can affect the time to failure of a pipe due to fatigue cracks, leading to significant uncertainty and conservation in determining integrity management and risk management strategies such as repair, pipe replacement, pressure reduction, and water pressure testing.

Disclosure of Invention

The invention aims to overcome the defects of the prior art in prediction accuracy and provide a method for greatly improving the prediction accuracy of the service life of a cracked pipeline.

The technical scheme adopted by the invention for solving the technical problems is as follows:

a method for predicting the service life of a cracked pipeline comprises the following steps:

step 1: simulating the pipeline crack by using a semiellipse, calculating a stress intensity factor of the surface of the pipeline crack, and selecting a Raju and Newman method or an API579 or BS7910 or a finite element simulation model to calculate stress concentration factors of all parts of the surface of the crack aiming at different pipeline strengths and physical properties;

step 2: establishing a crack growth model, establishing a growth model through two coupled Paris laws, and researching the fatigue expansion problem of the semi-elliptical surface cracks in the length and width expansion directions of the two cracks; the concrete formula is as follows:

in the formula, Δ K_AAnd Δ K_BA range of stress intensity factors at the surface point and depth of the surface crack; CA. CB, mA, and mB are material constants;

and step 3: compared with the unchanged model parameters of the traditional physical model, under a given m value, the growth prediction of the crack from the currently measured crack length and depth is carried out by using Paris law, and the crack size is expanded to the length and depth measured in the next inspection period in a certain period; observing the crack length and the crack depth in the next inspection period due to the existence of a measurement error e and a model error epsilon, namely obtaining the estimated crack length and depth, thereby obtaining a likelihood function l (a, b | m) for Bayesian inference of the following formula; according to the central limit theory, the influence of model errors on crack length estimation mainly depends on the mean value of the crack length estimation;

and 4, step 4: and predicting the crack growth after the data detection time point by using the updated model parameters to obtain a crack growth prediction model and a crack pipeline service life prediction model.

As a further improvement of the invention, C is_A=C_B、m_A=m_B(ii) a And (4) expanding the original hemielliptic crack into a new hemielliptic crack based on the two coupled Paris fatigue laws.

The invention has the beneficial effects that:

1. the prediction model method can be combined with a finite element modeling method or other empirical formulas, and integrates a crack growth physical model and online detection data to update model parameters by using a Bayesian method, so that more accurate life prediction is achieved.

Because the influence of the parameter m on the crack path and the prediction result is far greater than that of the parameter C, the more accurate life prediction can be obtained by updating the model parameter m.

Drawings

The invention is further illustrated with reference to the following figures and examples.

Fig. 1 is a crack propagation curve at m =3 for ILI #2 at 10 months and 18 days in 2013;

FIG. 2 is a crack propagation curve after physical model parameters are updated based on ILI #2 as a check point.

Detailed Description

The collected in-line inspection data (ILI) and non-destructive testing data (NDE) are shown in Table 1 below.

TABLE 1

Using the conventional physical model method, we used the on-line inspection data #1 as the basis of the crack initiation size data, and since the accuracy (standard deviation) of the on-line inspection tool was 0.0196 inch, the initial crack depth was estimated in the industry using a conservative method to be 0.039+0.0196=0.0586inch, and based on the crack propagation curve at m =3, the crack depth was 0.05923 inches at 10 months and 28 days in 2014, as shown in table 2 below. The new initial crack depths were 0.079+0.0196=0.0986 inches, respectively, as measured by ILI #2 at 10 months and 18 days in 2013. Finally, a crack propagation curve at m =3 for ILI #2 as shown in fig. 1 was obtained by a crack growth model, where the X-axis of fig. 1 is the aging time and the Y-axis is the crack depth.

TABLE 2

Using the proposed integration method under the same data conditions, our analysis results are shown below. We used ILI #2 as a checkpoint and the physical model parameter update results are shown in Table 3.

The crack depth predictions at 10 months and 28 days 2014 are shown in table 4. The analysis result can be observed, and the prediction precision of the existing physical model is greatly improved by the proposed method; meanwhile, fig. 2 is a crack propagation curve after updating of physical model parameters based on ILI #2 as a check point.

TABLE 3

TABLE 4

In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.

Claims

1. A method for predicting the service life of a crack pipeline is characterized by comprising the following steps: the method comprises the following steps:

step 2: establishing a crack growth model, establishing a growth model through two coupled Paris fatigue laws, and researching the fatigue expansion problem of the semielliptical surface cracks in the length and width expansion directions of the two cracks; the concrete formula is as follows:

in the formula,. DELTA.K_AAnd Δ K_BA range of stress intensity factors at the surface point and depth of the surface crack; c_A、C_B、m_AAnd m_BIs the material constant; da/dN and db/dN are the growth rates of the crack length and depth unit period, respectively; n is the number of cycles;

and step 3: compared with the unchanged model parameters of the traditional physical model, under the given m value, the growth prediction of the crack from the currently measured crack length and depth is carried out by using the Paris's law, and the crack size is expanded to the length and depth measured in the next inspection period under a certain period; observing the crack length and the crack depth in the next inspection period due to the existence of a measurement error e and a model error epsilon, so as to obtain the estimated crack length and depth, thereby obtaining a likelihood function l (a, b | m), and obtaining a post probability function by using a Bayesian method; according to the central limit theory, the influence of model errors on crack length estimation mainly depends on the mean value of the crack length estimation;

2. A method of predicting the life of a cracked pipe as claimed in claim 1, wherein: is provided with C_A＝C_B、m_A＝m_B(ii) a And (4) expanding the original hemielliptic crack into a new hemielliptic crack based on the two coupled Paris fatigue laws.