Pipeline failure probability prediction method and system and database construction method

Abstract

The invention relates to the field of pipeline systems, and discloses a pipeline failure probability prediction method and system and a database construction method. The invention discloses a pipeline failure probability prediction method. The method comprises the following steps: acquiring a pipeline failure database; acquiring working condition data of various working conditions of the actual pipeline, determining corresponding pipeline failure probabilities in the pipeline failure data of various working conditions according to the pipeline failure database, and calculating a basic failure probability under the condition of the working condition data of the actual pipeline; determining a probability correction coefficient according to the working condition data of the actual pipeline and the working condition data in the pipeline failure database; and correcting the basic failure probability according to the probability correction coefficient, and determining the predicted pipeline failure probability. The method has higher objectivity, accuracy and universality for predicting the failure probability of the pipeline.

Description

Technical Field

[0001] The present invention relates to the field of pipeline systems, and in particular to a pipeline failure probability prediction method, system and database construction method. Background Art

[0002] As the proportion of natural gas in the energy consumption structure increases, polyethylene (PE) is widely used in urban buried gas pipelines due to its corrosion resistance, simple installation and maintenance, good toughness and impact resistance, light weight, long service life, earthquake resistance and low comprehensive engineering cost, providing strong protection for people's daily life. At the same time, PE pipes still have disadvantages such as lower strength than steel pipes, easy deformation due to external forces, low pipeline bearing pressure and poor resistance to temperature changes. PE pipes are developing rapidly. More than 70% of the newly built urban gas pipelines are polyethylene pipes, and this proportion exceeds 90% of the newly laid medium and low pressure urban gas pipelines. The performance of PE pipes is increasingly recognized by the industry, and the safety issues caused by PE characteristics are also slowly emerging. However, due to large-scale urban construction and reconstruction, over-exploitation of groundwater resources, and the continuous increase in tall buildings, coupled with the impact of transportation facilities and transportation loads such as railways and bridges, the surface load has increased, the original mechanical equilibrium of the natural soil layer has been destroyed, and the ground settlement rate has continued to accelerate, causing geological settlement. As a result, the soil in the settlement area sinks or slides along the fracture surface under the action of gravity, generating normal pressure and tangential shear force on the buried gas pipeline below, ultimately leading to pipeline damage.

[0003] The existing technology mainly considers the probability of failure due to pipeline corrosion (such as CN108920421A or CN105302946A), but PE pipelines have strong corrosion resistance, and the failure probability prediction of such pipelines is not applicable to the failure probability prediction of PE pipelines. In addition, the patent application with publication number CN112183912A discloses a method and device for evaluating the failure probability of oil station pipelines, which obtains the basic failure probability of the pipeline according to the internal risk factors of the oil station pipeline; obtains the failure correction coefficient according to the external risk factors of the oil station pipeline; and obtains the failure probability of the pipeline according to the basic failure probability of the pipeline and the failure correction coefficient. This scheme mainly uses quantitative scoring to determine the data of each factor, which is highly subjective and the accuracy cannot be guaranteed, and thus easily leads to inaccurate final prediction results. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a pipeline failure probability prediction method, system and database construction method with higher universality and accuracy.

[0005] The pipeline failure probability prediction method disclosed in the present invention comprises the following steps:

[0006] Acquire a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions includes working condition data and pipeline failure probability corresponding to the working condition data;

[0007] Collect the working data of various working conditions of the actual pipeline, determine the corresponding pipeline failure probability in the pipeline failure data of various working conditions according to the pipeline failure database, and calculate the basic failure probability under the working data conditions of the actual pipeline;

[0008] Determine the probability correction coefficient based on the actual pipeline operating data and the operating data in the pipeline failure database;

[0009] The basic failure probability is corrected with the probability correction coefficient to determine the predicted pipeline failure probability.

[0010] Preferably, the pipeline failure database is obtained by construction, and the construction method of the pipeline failure database comprises the following steps:

[0011] Identify various operating conditions;

[0012] Based on the determined working conditions, original working condition data and corresponding failure information are respectively input;

[0013] Determine respectively under various single operating data conditions the critical value of the pipeline being determined not to have failed as the first target parameter value and the critical value of the pipeline being determined to have failed as the second target parameter value;

[0014] A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

[0015] Preferably, before determining the first target parameter value and the second target parameter value, various original operating condition data are arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original operating condition data arranged in sequence.

[0016] Preferably, when determining the failure probability corresponding to each operating condition data interval, the number of original operating condition data of pipeline failure in the operating condition data interval is counted and divided by the total number of original operating condition data in the operating condition data interval to obtain the failure probability corresponding to the operating condition data interval.

[0017] Preferably, the pipeline failure database contains pipeline failure data of at least three working conditions, and when collecting working condition data of the actual pipeline, working condition data of the three working conditions are collected.

[0018] Preferably, the pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data;

[0019] When collecting the actual working condition data of the pipeline, the tensile stress, bending stress and residual stress of the pipeline are collected.

[0020] Preferably, the tensile stress is a uniformly distributed tensile stress caused by the internal pressure of the pipeline on the pipe wall;

[0021] The bending stress is the bending stress caused by the outer surface of the pipeline being compressed and the inner surface being tensile, which is formed when the pipeline is subjected to the soil load and traffic load.

[0022] The residual stress is the residual stress caused by the different cooling rates of the inner and outer walls of the pipeline, resulting in the inner surface being tensile and the outer surface being compressive.

[0023] Preferably, the pipeline failure database includes working condition data intervals of various working conditions and pipeline failure probabilities corresponding to the working condition data intervals. The basic failure probability is calculated as follows:

[0024] Determine the actual tensile stress σ P The interval (σ Pj1 , σ Pd1 ) and the corresponding failure probability p jd1 , σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval;

[0025] Determine the actual bending stress σ F The interval (σ Fj2 , σ Fd2 ) and the corresponding failure probability p jd2 , σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval;

[0026] Determine the residual stress σ R The interval (σ Rj3 , σ Rd3 ) and the corresponding failure probability p jd3 , σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval;

[0027] The fuzzy set that determines the fuzzy event is:

[0028]

[0029] μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval;

[0030]

[0031]

[0032]

[0033] The probability of pipeline failure under the action of three stresses is determined as:

[0034] p(A)=μ(σ P ) jd1 +μ(σ F ) jd2 +μ(σ R ) jd3

[0035] p(A) represents the basic failure probability of the pipeline.

[0036] Preferably, when determining the probability correction coefficient, the Euclidean distance between the actual pipeline operating condition data and the critical value of each operating condition in the pipeline failure database is calculated, and the probability correction coefficient is determined according to the Euclidean distance.

[0037] Preferably, when determining the probability correction coefficient, the Euclidean distance between the actual pipeline operating condition data and the critical value of each operating condition in the pipeline failure database is calculated, and the probability correction coefficient is determined according to the Euclidean distance;

[0038] According to the first target tensile stress σ Pmin , the second target tensile stress σ Pmax , the first target bending stress σ Fmin , the second target bending stress σ Fmax , the first target residual stress σ Rmin And the second target residual stress σ Rmax , determine the probability correction coefficient p' as:

[0039]

[0040] in,

[0041]

[0042] α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

[0043] Preferably, when the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%;

[0044] When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R are respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%;

[0045] The first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

[0046] Preferably, the basic failure probability is corrected by the probability correction coefficient to determine the failure probability of the PE pipeline as follows:

[0047]

[0048] The pipeline failure probability prediction device disclosed in the present application includes:

[0049] A database unit stores a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions include working condition data and pipeline failure probability corresponding to the working condition data;

[0050] An input unit, used to input working condition data of various working conditions of actual pipelines;

[0051] A matching unit, for determining corresponding pipeline failure probabilities in pipeline failure data of various working conditions according to input working condition data of various working conditions of actual pipelines and according to a pipeline failure database;

[0052] A basic failure probability calculation unit calculates the basic failure probability under the actual pipeline working condition data conditions according to the pipeline failure probability determined in the pipeline failure data of various working conditions;

[0053] A probability correction coefficient calculation unit, used to determine the probability correction coefficient according to the actual pipeline working condition data and the working condition data in the pipeline failure database;

[0054] A pipeline failure probability calculation unit, used to calculate the predicted pipeline failure probability based on the basic failure probability and the probability correction coefficient;

[0055] The output unit is used to output the predicted pipeline failure probability.

[0056] Preferably, the database unit has a tensile stress pipeline failure data subunit, a bending stress pipeline failure data subunit and a residual stress pipeline failure data subunit.

[0057] Preferably, the pipeline failure database includes working condition data intervals of various working conditions and pipeline failure probabilities corresponding to the working condition data intervals;

[0058] The basic failure probability calculation formula of the pipeline of the basic failure probability calculation unit is as follows:

[0059] p(A)=μ(σ P ) jd1 +μ(σ F ) jd2 +μ(σ R ) jd3

[0060] in, μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval;

[0061]

[0062]

[0063]

[0064] σ P represents the actual tensile stress, σ F represents the actual bending stress, σ R represents the residual stress, σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval, σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval, σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval.

[0065] Preferably, the probability correction coefficient p' of the probability correction coefficient calculation unit is:

[0066]

[0067] in,

[0068]

[0069] α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ RmaxIndicates the critical value of residual stress that determines the failure of the pipeline.

[0070] Preferably, the pipeline failure probability calculation unit includes a direct judgment unit,

[0071] When the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%;

[0072] When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R Respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%.

[0073] Preferably, the failure probability calculation formula of the pipeline failure probability calculation unit is:

[0074] The pipeline failure database construction method disclosed in the present application comprises the following steps:

[0075] Identify various operating conditions;

[0076] Input the original working condition data and the corresponding failure information;

[0077] Determine respectively under various single operating data conditions the critical value of the pipeline being determined not to have failed as the first target parameter value and the critical value of the pipeline being determined to have failed as the second target parameter value;

[0078] A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

[0079] Preferably, before determining the first target parameter value and the second target parameter value, various original operating condition data are arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original operating condition data arranged in sequence.

[0080] Preferably, when determining the failure probability corresponding to each operating condition data interval, the number of original operating condition data of pipeline failure in the operating condition data interval is counted and divided by the total number of original operating condition data in the operating condition data interval to obtain the failure probability corresponding to the operating condition data interval.

[0081] Preferably, the pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data.

[0082] The beneficial effects of the present invention are as follows: the present application predicts the failure probability of the pipeline according to the actual environment in which the pipeline is located, and can also simulate the service environment of the pipeline to predict whether there is a risk of failure after service, so that the staff can make timely adjustments to ensure the normal service of the pipeline. Compared with the prediction method based on multi-factor quantitative scoring, the present application has higher objectivity and accuracy, and has higher universality than the existing prediction method that considers failure caused by pipeline corrosion, and can be applied to the failure prediction of PE pipelines and other pipelines. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a flowchart of an embodiment of the present application. DETAILED DESCRIPTION

[0084] The present invention is further described below.

[0085] The pipeline failure probability prediction method disclosed in the present invention comprises the following steps:

[0086] Acquire a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions includes working condition data and pipeline failure probability corresponding to the working condition data;

[0087] Collect the working data of various working conditions of the actual pipeline, determine the corresponding pipeline failure probability in the pipeline failure data of various working conditions according to the pipeline failure database, and calculate the basic failure probability under the working data conditions of the actual pipeline;

[0088] Determine the probability correction coefficient based on the actual pipeline operating data and the operating data in the pipeline failure database;

[0089] The basic failure probability is corrected with the probability correction coefficient to determine the predicted pipeline failure probability.

[0090] The pipeline failure database is used to store pipeline failure data of various working conditions. The collected working condition data of various working conditions of the actual pipeline needs to be the working condition data contained in the pipeline failure database. For example, the pipeline failure data contains tensile stress, bending stress and residual stress. The tensile stress, bending stress and residual stress of the actual pipeline can be collected. Of course, the pipeline failure data can also include more types of working condition data, such as the reaction force of the pipeline support, the wind load of the overhead pipeline, etc. When collecting the working condition data of the actual pipeline, only public data with obvious influence is used.

[0091] Based on the working condition data of various working conditions of the actual pipeline, the corresponding pipeline failure probability can be determined in the pipeline failure data of various working conditions in the pipeline failure database, and the pipeline failure probabilities of these single working conditions can be combined to calculate the pipeline failure probability under various working conditions as the basic failure probability. Due to the different degrees of influence of various working conditions, the basic failure probability cannot accurately express the failure probability of the actual pipeline. Therefore, this application determines the probability correction coefficient based on the working condition data of the actual pipeline and the working condition data in the pipeline failure database, and corrects the basic failure probability with the probability correction coefficient to determine the predicted pipeline failure probability.

[0092] The pipeline failure database is the basis and core of the present application to achieve pipeline failure probability prediction. The pipeline failure database can be self-built or directly purchased as a mature database. Here, the present application also provides a method for building the pipeline failure database, which includes the following steps:

[0093] Identify various operating conditions;

[0094] Based on the determined working conditions, original working condition data and corresponding failure information are respectively input;

[0095] Determine respectively under various single operating data conditions the critical value at which the pipeline is determined not to fail as the first target parameter value and the critical value at which the pipeline is determined to fail as the second target parameter value;

[0096] A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

[0097] The database is established so that the failure probability can be predicted directly based on the database during subsequent use. The type of database working condition can be confirmed according to the predicted pipeline object. Usually, the working condition that has a significant impact on pipeline failure is used as the working condition of the pipeline failure database, and then the original working condition data and the corresponding failure information are input. Since this scheme uses frequency instead of probability as the basic probability, the more data in the database, the better, such as: 10,000 sets of data, and the original working condition data should have sufficient coverage, so that the data between the pipeline must not fail and the stress must fail under different working conditions exist, so as to determine the first target parameter value and the second target parameter value later.

[0098] Each working condition includes multiple sets of experimental data, such as: there are different tensile stresses σ P1 , σ P2 ..., each tensile stress data has a corresponding pipeline failure state, and the failure state includes failure or non-failure. The failure state of the pipeline is failure or non-failure. Failure usually indicates cracks or other damage, and non-failure indicates that the pipeline is in a normal state. The same operating condition data can appear multiple times in the same operating condition. For example, although they are two identical operating condition data, they correspond to the two states of pipeline failure and non-failure respectively. Through the accumulation of a large amount of data, the corresponding failure probability for a certain operating condition data can be determined. The failure probability is determined by a large amount of data and has better reference.

[0099] When collecting data, several operating condition data should be collected separately. For example, tensile stress should be collected separately, and bending stress and residual stress should be controlled to the minimum. It can be assumed that under the action of this set of standard forces, the pipeline will not fail through multiple experiments, and then a single force can be changed to determine the impact of a single operating condition data on the pipeline.

[0100] The first target parameter value is the critical value for determining that the pipeline has not failed under the condition of single operating data, and the second target parameter value is the critical value for determining that the pipeline has failed under the condition of single operating data. The failure probability of the first target parameter value is 0%, and each operating data thereafter must have a failure state, so the operating data interval with a pipeline failure probability of 0 can be determined. The failure probability of the second target parameter value is 100%, and each operating data before it must have a non-failure state, so the operating data interval with a pipeline failure probability of 100% can be determined.

[0101] Since the failure information of a single working condition data has a certain degree of randomness, in order to reduce the randomness, a plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data and the corresponding failure information in each working condition data interval. As for the calculation of the failure probability, in a preferred embodiment of the present application, when determining the failure probability corresponding to each working condition data interval, the number of original working condition data of pipeline failure in the working condition data interval is counted, and divided by the total number of original working condition data in the working condition data interval to obtain the failure probability corresponding to the working condition data interval. For example, if there are 50 failures and 50 non-failures in the original working condition data in a certain interval, the failure probability of the working condition data interval is 50%. The specific number of working condition data intervals is determined according to the original working condition data. Generally, the larger the amount of original working condition data, the more data in the working condition data interval can be divided, and the probability calculation is more accurate.

[0102] The first target parameter value and the second target parameter value can be determined by experienced experts or other existing databases, or can be determined directly based on the input original working condition data. In the latter method, in order to facilitate the determination of the first target parameter value and the second target parameter value, as a preferred embodiment of the present application, before determining the first target parameter value and the second target parameter value, various original working condition data are arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original working condition data arranged in order. In addition, for the case where the amount of data is relatively small, the first target parameter value can also be the previous value of a certain working condition data with the smallest failure probability in the database.

[0103] The total number of working conditions is determined according to the impact of the actual working conditions on the pipeline. Usually, working conditions that have a significant impact on pipeline failure are screened out. In a preferred embodiment of the present application, the pipeline failure database has pipeline failure data of at least three working conditions. When collecting working condition data of the actual pipeline, working condition data of the three working conditions are collected. The number of working conditions in the pipeline failure database can be higher than the number of working conditions actually collected. When predicting different pipeline failure probabilities, appropriate types of working conditions are selected for collection.

[0104] In a preferred embodiment of the present application, the pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data; when collecting the actual pipeline working condition data, the tensile stress, bending stress and residual stress of the pipeline are collected. Of course, the stress can also be replaced by the other three influencing parameters to predict the probability of pipeline failure under other parameters. For example, when the pipeline is not subjected to the soil cover load and the temperature inside and outside the pipe is relatively balanced, its bending stress and residual stress are relatively small, and both can be replaced by other influencing parameters.

[0105] Since the tensile stress, bending stress and residual stress of the actual pipeline are not easy to measure directly, therefore, in the preferred embodiment of the present application, the tensile stress is the uniformly distributed tensile stress caused by the internal pressure of the pipeline on the pipe wall; the bending stress is the bending stress formed by the outer surface of the pipeline being compressed and the inner surface being tensile due to the overburden load and traffic load on the pipeline; the residual stress is the residual stress formed by the inner surface being tensile and the outer surface being compressed due to the different cooling rates of the inner and outer walls of the pipeline, and is specifically calculated as follows:

[0106] Determine the relatively uniform tensile stress σ caused by the internal pressure of the pipeline on the pipe wall P for:

[0107]

[0108] Among them, P represents the internal pressure of the pipeline, D represents the diameter of the pipeline, and t represents the wall thickness of the pipeline.

[0109] Secondly, determine the bending stress σ caused by the pipe's outer surface being compressed and its inner surface being tensile due to the soil load and traffic load. F for:

[0110]

[0111]

[0112] Among them, K m represents the plane bending moment coefficient, W represents the total vertical load, E p represents the elastic modulus of the pipe, r represents the pipe radius of the pipe, K d represents the internal pressure coefficient, E' represents the elastic modulus of the soil, C d represents the load calculation coefficient, γ' represents the effective bulk density of the soil, B d Indicates the calculated width of the groove, I c Indicates the traffic load impact coefficient, C t represents the surface load factor, F represents the wheel load, and L represents the effective calculated length of the pipeline.

[0113] Finally, determine the residual stress σ caused by the different cooling rates of the inner and outer walls of the pipeline, which results in the inner surface being tensile and the outer surface being compressive. R for:

[0114]

[0115] Among them, σ c represents the maximum residual stress of the pipeline, r t It represents the radial distance from the inner surface of the pipe to the inner wall of the pipe, which can be a pre-determined parameter.

[0116] Taking the above-mentioned tensile stress, bending stress and residual stress as the actual collected working conditions as an example, the pipeline failure database contains working condition data intervals of various working conditions and pipeline failure probabilities of corresponding working condition data intervals. The basic failure probability is calculated as follows:

[0117] Determine the actual tensile stress σ P The interval (σ Pj1 , σ Pd1 ) and the corresponding failure probability p jd1 , σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval;

[0118] Determine the actual bending stress σ F The interval (σ Fj2 , σ Fd2 ) and the corresponding failure probability p jd2 , σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval;

[0119] Determine the residual stress σ R The interval (σ Rj3 , σ Rd3 ) and the corresponding failure probability p jd3 , σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval;

[0120] The fuzzy set that determines the fuzzy event is:

[0121]

[0122] μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval;

[0123]

[0124]

[0125]

[0126] The probability of pipeline failure under the action of three stresses is determined as:

[0127] p(A)=μ(σ P ) jd1 +μ(σ F ) jd2 +μ(σ R ) jd3

[0128] p(A) represents the basic failure probability of the pipeline.

[0129] Since the degree of influence of each working condition is different, the present application uses a probability correction coefficient to correct the basic failure probability in order to obtain a more accurate prediction result. The probability correction coefficient can be confirmed through experience or past failure statistics. In a preferred embodiment of the present application, when determining the probability correction coefficient, the Euclidean distance between the working condition data of the actual pipeline and the critical value of each working condition in the pipeline failure database is calculated, and the probability correction coefficient is determined based on the Euclidean distance. In this way, the influence of each working condition on pipeline failure can be directly quantified and calculated, which not only obtains more accurate results, but also makes the calculation process simpler.

[0130] Taking tensile stress, bending stress and residual stress as examples, when determining the probability correction coefficient, the Euclidean distance between the actual pipeline operating condition data and the critical value of each operating condition in the pipeline failure database is calculated, and the probability correction coefficient is determined based on the Euclidean distance;

[0131] According to the first target tensile stress σ Pmin , the second target tensile stress σ Pmax , the first target bending stress σ Fmin , the second target bending stress σ Fmax , the first target residual stress σ Rmin And the second target residual stress σ Rmax , determine the probability correction coefficient p' as:

[0132]

[0133] in,

[0134]

[0135] α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σFmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

[0136] It is worth noting that when the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%;

[0137] When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R are respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%;

[0138] The first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

[0139] Furthermore, the basic failure probability is corrected by the probability correction coefficient to determine the failure probability of the PE pipeline:

[0140]

[0141] The embodiment of the present application can be divided into several units according to the functions of each step of the pipeline failure probability prediction method. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated module can be implemented in the form of hardware or software.

[0142] Based on this, the present application also provides a pipeline failure probability prediction device, comprising:

[0143] A database unit stores a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions include working condition data and pipeline failure probability corresponding to the working condition data;

[0144] An input unit, used to input working condition data of various working conditions of actual pipelines;

[0145] A matching unit, for determining corresponding pipeline failure probabilities in pipeline failure data of various working conditions according to input working condition data of various working conditions of actual pipelines and according to a pipeline failure database;

[0146] A basic failure probability calculation unit calculates the basic failure probability under the actual pipeline working condition data conditions according to the pipeline failure probability determined in the pipeline failure data of various working conditions;

[0147] A probability correction coefficient calculation unit, used to determine the probability correction coefficient according to the actual pipeline working condition data and the working condition data in the pipeline failure database;

[0148] A pipeline failure probability calculation unit, used to calculate the predicted pipeline failure probability based on the basic failure probability and the probability correction coefficient;

[0149] The output unit is used to output the predicted pipeline failure probability.

[0150] The database unit is a device for storing the pipeline failure database, which can be an online database or a local database. The input unit can use an existing physical keyboard, a virtual keyboard, voice input, etc. The matching unit, the basic failure probability calculation unit, the probability correction coefficient calculation unit, and the pipeline failure probability calculation unit can all be implemented in the form of software. The output unit usually uses existing devices such as a display and voice broadcast.

[0151] In a preferred embodiment of the present application, the database unit has a tensile stress pipeline failure data subunit, a bending stress pipeline failure data subunit and a residual stress pipeline failure data subunit.

[0152] In a preferred embodiment of the present application, the pipeline failure database includes working condition data intervals of various working conditions and pipeline failure probabilities corresponding to the working condition data intervals;

[0153] The basic failure probability calculation formula of the pipeline of the basic failure probability calculation unit is as follows:

[0154] p(A)=μ(σ P )jd1 +μ(σ F ) jd2 +μ(σ R ) jd3

[0155] in, μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval;

[0156]

[0157]

[0158]

[0159] σ P represents the actual tensile stress, σ F represents the actual bending stress, σ R represents the residual stress, σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval, σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval, σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval.

[0160] In a preferred embodiment of the present application, the probability correction coefficient p' of the probability correction coefficient calculation unit is:

[0161]

[0162] in,

[0163]

[0164] α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fminrepresents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

[0165] In a preferred embodiment of the present application, the pipeline failure probability calculation unit includes a direct judgment unit,

[0166] When the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%;

[0167] When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R are respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%.

[0168] In a preferred embodiment of the present application, the failure probability calculation formula of the pipeline failure probability calculation unit is:

[0169]

[0170] The various functional units in the pipeline failure probability prediction device have been described in detail in the aforementioned method, and will not be repeated here.

[0171] As mentioned above, the pipeline failure database is the basis of this application. Regarding the components of the database, this application also provides a pipeline failure database construction method, including the following steps:

[0172] Identify various operating conditions;

[0173] Input the original working condition data and the corresponding failure information;

[0174] Determine respectively under various single operating data conditions the critical value of the pipeline being determined not to have failed as the first target parameter value and the critical value of the pipeline being determined to have failed as the second target parameter value;

[0175] A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

[0176] Taking PE pipes as an example, because PE pipes are extremely corrosion-resistant, there is no need to consider the corrosion effect of the environment on PE pipes. Therefore, without considering human damage, the main forces on PE pipes come from the tensile stress caused by the internal pressure of the pipe, the bending stress caused by the pipe's soil load and traffic load, and the residual stress caused by the temperature difference between the inside and outside. In the database, the failure state of PE pipes is failure or non-failure. Failure can mean that cracks appear in the PE pipe, and non-failure can mean that there are no cracks in the PE pipe.

[0177] The same tensile stress data can appear multiple times, as can bending stress and residual stress. By accumulating a large amount of data, the corresponding failure probability at a certain tensile stress data can be determined. This failure probability is determined by a large amount of data and has a certain reference value. The same is true for bending stress and residual stress.

[0178] When collecting data, several types of data should be collected separately. For example, tensile stress should be collected separately, and bending stress and residual stress should be controlled to the minimum. It can be assumed that under the action of this set of standard forces, through multiple experiments, the PE pipeline will definitely not fail, and then a single force can be changed to determine the impact of a single stress on the PE pipeline.

[0179] Similar to what was described above, in a preferred embodiment of the present application, before determining the first target parameter value and the second target parameter value, various original operating condition data are first arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original operating condition data arranged in sequence.

[0180] In a preferred embodiment of the present application, when determining the failure probability corresponding to each operating condition data interval, the number of original operating condition data of pipeline failure in the operating condition data interval is counted and divided by the total number of original operating condition data in the operating condition data interval to obtain the failure probability corresponding to the operating condition data interval.

[0181] In a preferred embodiment of the present application, the pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data.

[0182] In actual application, since the construction of the database requires huge costs, the provider and user of the pipeline failure database can be different units. For example, the database provider can contain a variety of working conditions in the database, while the user who actually predicts pipeline failure does not need to purchase the database service directly from the database provider. The database provider can provide services to multiple users at the same time.

[0183] like Figure 1 As shown, the overall process of this application is briefly described below using PE pipe as an example.

[0184] In the pipeline failure database, three types of stress, namely tensile stress, bending stress and residual stress, are identified as input working conditions. Each working condition includes multiple sets of experimental data, such as: there are different tensile stresses σ P1 , σ P2 ..., each tensile stress data has a corresponding PE pipe failure state, the failure state includes failure or non-failure. There are different bending stresses σ F1 , σ F2 ..., each bending stress data has a corresponding PE pipeline failure state. There are different residual stresses σ R1 , σ R2 ..., each residual stress data has a corresponding PE pipeline failure state.

[0185] A first target tensile stress, a second target tensile stress, a first target bending stress, a second target bending stress, a first target residual stress and a second target residual stress are determined as failure critical values ​​in the pipeline failure database.

[0186] The working condition data between the first target tensile stress and the second target tensile stress are divided into N spaces. The failure probability of each interval is calculated by dividing the number of tensile stress data corresponding to the interval with the failure state by the total number of tensile stress data in the interval. The same treatment is done for bending stress and residual stress. The construction of the pipeline failure database can provide data support for subsequent predictions.

[0187] When predicting the actual pipeline, the actual tensile stress, actual bending stress and actual residual stress are collected, and the actual tensile stress data interval, the actual bending stress data interval and the actual residual stress data interval are determined. The membership and failure probability of the actual tensile stress in the corresponding tensile stress data interval are determined, the membership and failure probability of the actual bending stress in the corresponding bending stress interval are determined, and the membership and failure probability of the actual residual stress in the corresponding residual stress interval are determined.

[0188] Based on the membership degree and failure probability of the previous step, the basic failure probability of the PE pipeline under various influencing factors is determined, and the probability correction coefficient of the basic failure probability is determined. According to the basic failure probability and the probability correction coefficient, the final failure probability of the PE pipeline is determined.

Claims

1. A pipeline failure probability prediction method, characterized in that: The steps include: Acquire a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions includes working condition data and pipeline failure probability corresponding to the working condition data; Collect the working data of various working conditions of the actual pipeline, determine the corresponding pipeline failure probability in the pipeline failure data of various working conditions according to the pipeline failure database, and calculate the basic failure probability under the working data conditions of the actual pipeline; Determine the probability correction coefficient based on the actual pipeline operating data and the operating data in the pipeline failure database; The basic failure probability is corrected with the probability correction coefficient to determine the predicted pipeline failure probability.

2. The pipeline failure probability prediction method according to claim 1, characterized in that: The pipeline failure database is obtained by construction, and the construction method of the pipeline failure database includes the following steps: Identify various operating conditions; Based on the determined working conditions, original working condition data and corresponding failure information are respectively input; Determine respectively under various single operating data conditions the critical value of the pipeline being determined not to have failed as the first target parameter value and the critical value of the pipeline being determined to have failed as the second target parameter value; A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

3. The pipeline failure probability prediction method according to claim 2, characterized in that: Before determining the first target parameter value and the second target parameter value, various original operating condition data are arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original operating condition data arranged in order.

4. The pipeline failure probability prediction method according to claim 2, characterized in that: When determining the failure probability corresponding to each operating condition data interval, the number of original operating condition data of pipeline failure in the operating condition data interval is counted and divided by the total number of original operating condition data in the operating condition data interval to obtain the failure probability corresponding to the operating condition data interval.

5. The pipeline failure probability prediction method according to any one of claims 1 to 4, characterized in that: The pipeline failure database contains pipeline failure data of at least three working conditions. When collecting working condition data of actual pipelines, working condition data of the three working conditions are collected.

6. The pipeline failure probability prediction method according to claim 5, characterized in that: The pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data; When collecting the actual working condition data of the pipeline, the tensile stress, bending stress and residual stress of the pipeline are collected.

7. The pipeline failure probability prediction method according to claim 6, characterized in that: The tensile stress is the uniformly distributed tensile stress caused by the internal pressure of the pipeline on the pipe wall; The bending stress is the bending stress caused by the outer surface of the pipeline being compressed and the inner surface being tensile, which is formed when the pipeline is subjected to the soil load and traffic load. The residual stress is the residual stress caused by the different cooling rates of the inner and outer walls of the pipeline, resulting in the inner surface being tensile and the outer surface being compressive.

8. The pipeline failure probability prediction method according to claim 6, characterized in that: The pipeline failure database includes working condition data intervals of various working conditions and pipeline failure probabilities corresponding to the working condition data intervals. The basic failure probability is calculated as follows: Determine the actual tensile stress σ P The interval (σ Pj1 , σ Pd1 ) and the corresponding failure probability p jd1 , σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval; Determine the actual bending stress σ F The interval (σ Fj2 , σ Fd2 ) and the corresponding failure probability p jd2 , σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval; Determine the residual stress σ R The interval (σ Rj3 , σ Rd3 ) and the corresponding failure probability p jd3 , σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval; The fuzzy set that determines the fuzzy event is: μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval; The probability of pipeline failure under the action of three stresses is determined as: p(A)=μ(σ P )p jd1 +μ(σ F )p jd2 +μ(σ R )p jd3 p(A) represents the basic failure probability of the pipeline.

9. The pipeline failure probability prediction method according to claim 1, characterized in that: When determining the probability correction coefficient, the Euclidean distance between the actual pipeline operating condition data and the critical value of each operating condition in the pipeline failure database is calculated, and the probability correction coefficient is determined based on the Euclidean distance.

10. The pipeline failure probability prediction method according to claim 8, characterized in that: When determining the probability correction coefficient, the Euclidean distance between the actual pipeline operating condition data and the critical value of each operating condition in the pipeline failure database is calculated, and the probability correction coefficient is determined according to the Euclidean distance; According to the first target tensile stress σ Pmin , the second target tensile stress σ Pmax , the first target bending stress σ Fmin , the second target bending stress σ Fmax , the first target residual stress σ Rmin And the second target residual stress σ Rmax , determine the probability correction coefficient p' as: in, α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

11. The pipeline failure probability prediction method according to claim 8 or 10, characterized in that: When the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%; When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R are respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%; The first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

12. The pipeline failure probability prediction method according to claim 10, characterized in that: The basic failure probability is corrected by the probability correction coefficient, and the failure probability of the PE pipeline is determined as:

13. A pipeline failure probability prediction device, characterized in that: include: A database unit stores a pipeline failure database, wherein the pipeline failure database stores pipeline failure data of various working conditions, wherein the pipeline failure data of various working conditions include working condition data and pipeline failure probability corresponding to the working condition data; An input unit, used to input working condition data of various working conditions of actual pipelines; A matching unit, for determining corresponding pipeline failure probabilities in pipeline failure data of various working conditions according to input working condition data of various working conditions of actual pipelines and according to a pipeline failure database; A basic failure probability calculation unit calculates the basic failure probability under the actual pipeline working condition data conditions according to the pipeline failure probability determined in the pipeline failure data of various working conditions; A probability correction coefficient calculation unit, used to determine the probability correction coefficient according to the actual pipeline working condition data and the working condition data in the pipeline failure database; A pipeline failure probability calculation unit, used to calculate the predicted pipeline failure probability based on the basic failure probability and the probability correction coefficient; The output unit is used to output the predicted pipeline failure probability.

14. The pipeline failure probability prediction device according to claim 13, characterized in that: The database unit comprises a tensile stress pipeline failure data subunit, a bending stress pipeline failure data subunit and a residual stress pipeline failure data subunit.

15. The pipeline failure probability prediction device according to claim 14, characterized in that: The pipeline failure database includes working condition data intervals of various working conditions and pipeline failure probabilities corresponding to the working condition data intervals; The basic failure probability calculation formula of the pipeline of the basic failure probability calculation unit is as follows: p(A)=μ(σ P )p jd1 +μ(σ F )p jd2 +μ(σ R )p jd3 in, μ(σ P ) represents the actual tensile stress σ of the pipeline P The membership degree of the corresponding interval, μ(σ F ) represents the actual bending stress σ of the pipe F The membership degree of the corresponding interval, μ(σ R ) represents the actual residual stress σ R The membership degree of the corresponding interval; σ P represents the actual tensile stress, σ F represents the actual bending stress, σ R represents the residual stress, σ Pj1 , σ Pd1 Represent the actual tensile stress σ P The upper and lower limits of the stress interval, σ Fj2 , σ Fd2 They represent the actual bending stress σ F The upper and lower limits of the stress interval, σ Rj3 , σ Rd3 Respectively represent the residual stress σ R The upper and lower limits of the stress interval.

16. The pipeline failure probability prediction device according to claim 15, characterized in that: Probability correction coefficient calculation unit probability correction coefficient p' is: in, α represents a correction factor set to a constant; the first target tensile stress σ Pmin represents the critical tensile stress value of the pipeline to determine whether it has not failed, and the second target tensile stress σ Pmax represents the critical value of tensile stress for pipeline failure, the first target bending stress σ Fmin represents the critical bending stress value of the pipeline to determine whether it has not failed, and the second target bending stress σ Fmax It represents the critical value of bending stress for determining failure of the pipeline, and the first target residual stress σ Rmin It represents the critical value of residual stress of the pipeline to determine whether it has not failed, and the second target residual stress σ Rmax Indicates the critical value of residual stress that determines the failure of the pipeline.

17. The pipeline failure probability prediction device according to claim 16, characterized in that: The pipeline failure probability calculation unit includes a direct judgment unit, When the actual tensile stress σ P , actual bending stress σ F And the actual residual stress σ R are respectively less than the first target tensile stress σ Pmin , the first target bending stress σ Fmin and the first target residual stress σ Rmin When , the pipeline failure probability is directly judged to be 0%; When the actual tensile stress σ P , actual bending stress σ F Or the actual residual stress σ R are respectively greater than the second target tensile stress σ Pmax , the second target bending stress σ Fmax And the second target residual stress σ Rmax When , it can be directly judged that the failure probability of PE pipeline is 100%.

18. The pipeline failure probability prediction device according to claim 16, characterized in that: The failure probability calculation formula of the pipeline failure probability calculation unit is:

19. A pipeline failure database construction method, characterized in that: The steps include: Identify various operating conditions; Input the original working condition data and the corresponding failure information; Determine respectively under various single operating data conditions the critical value of the pipeline being determined not to have failed as the first target parameter value and the critical value of the pipeline being determined to have failed as the second target parameter value; A plurality of working condition data intervals are divided between the first target parameter value and the second target parameter value, and the failure probability corresponding to each working condition data interval is determined according to the original working condition data in each working condition data interval and the corresponding failure information. The working condition data interval with a pipeline failure probability of 0 is determined based on the first target parameter value, and the working condition data interval with a pipeline failure probability of 100% is determined based on the second target parameter value, so as to obtain pipeline failure data of various working conditions, and the pipeline failure data of various working conditions all include the working condition data interval and the pipeline failure probability of the corresponding working condition data interval.

20. The pipeline failure database construction method according to claim 19, characterized in that: Before determining the first target parameter value and the second target parameter value, various original operating condition data are arranged in order from small to large or from large to small, and the first target parameter value and the second target parameter value are determined from the various original operating condition data arranged in order.

21. The pipeline failure database construction method according to claim 19, characterized in that: When determining the failure probability corresponding to each operating condition data interval, the number of original operating condition data of pipeline failure in the operating condition data interval is counted and divided by the total number of original operating condition data in the operating condition data interval to obtain the failure probability corresponding to the operating condition data interval.

22. The pipeline failure database construction method according to claim 19, characterized in that: The pipeline failure database includes tensile stress pipeline failure data, bending stress pipeline failure data and residual stress pipeline failure data.

Images