A method, device, equipment and medium for ranking the risk of girth welds of in-service pipelines

Abstract

This invention relates to a method, apparatus, equipment, and medium for risk ranking of circumferential welds in in-service pipelines. The method includes: acquiring multiple risk factors for the welds to be ranked; for each risk factor, acquiring the weight corresponding to the risk factor and the level score of the risk factor under a specified attribute feature; for each risk factor, determining the comprehensive risk value of the risk factor based on the weight and level score; and ranking the risks of the pipeline circumferential welds based on the comprehensive risk values ​​of each risk factor. The method of this invention can be used for risk ranking of welds in in-service gas transmission pipelines, guiding the current domestic circumferential weld inspection, effectively reducing excavation volume, and improving excavation accuracy.

Description

Technical Field

[0001] This invention relates to the field of oil and gas pipeline circumferential weld integrity management technology. Specifically, this invention relates to a method, apparatus, equipment, and medium for risk ranking of circumferential welds in in-service pipelines. Background Technology

[0002] In recent years, the failure of circumferential welds in long-distance oil and gas pipelines has become a major accident affecting pipeline safety. In order to effectively manage the risks of circumferential welds, domestic pipeline companies have organized and carried out circumferential weld quality risk investigation work, discovered a large number of unqualified welds and cracked welds, and accumulated a large amount of data and results of excavated welds.

[0003] These data and results were statistically analyzed manually, revealing some patterns and clarifying some issues. However, due to the complexity of the causes of circumferential weld defects and the correlation between various influencing factors, manual analysis could not draw definitive conclusions and could not effectively guide circumferential weld excavation work, develop effective weld inspection plans, or significantly reduce the number of weld excavations and inspections required.

[0004] Data mining is an effective method for solving the correlation of large amounts of data and finding patterns. In-depth mining and mathematical statistical analysis of circumferential weld inspection data are crucial for establishing a risk ranking model for gas pipeline circumferential welds and developing a circumferential weld risk ranking technology. Improving the accuracy of excavation of cracked weld joints in in-service high-strength steel pipelines is a pressing task. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, apparatus, equipment and medium for risk ranking of circumferential welds in in-service pipelines, aiming to solve at least one of the above-mentioned technical problems.

[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A risk ranking method for circumferential welds of in-service pipelines, the method comprising:

[0007] Obtain multiple risk factors for welds to be prioritized;

[0008] For each risk factor, obtain the weight corresponding to the risk factor and the level score corresponding to the risk factor under the specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor.

[0009] For each of the aforementioned risk factors, a comprehensive risk value for that risk factor is determined based on its weight and level score.

[0010] Based on the comprehensive risk value of each of the aforementioned risk factors, the risks of pipeline circumferential welds are ranked.

[0011] The beneficial effects of this invention are as follows: In this application, multiple risk factors affecting the risk of pipeline welds are considered, as well as the degree of risk of each risk factor relative to the other risk factors, and the degree of risk of the specified attribute characteristics of each risk factor relative to the other attribute characteristics of that risk factor. The comprehensive risk value of each risk factor is determined from multiple aspects. Finally, based on the comprehensive risk value of each risk factor, the risk of pipeline circumferential welds is ranked. Through the above scheme, the current domestic circumferential weld inspection can be guided, effectively reducing the amount of excavation and improving the excavation accuracy.

[0012] Based on the above technical solution, the present invention can be further improved as follows.

[0013] Furthermore, for each of the aforementioned risk factors, obtaining the weight corresponding to the risk factor and the sum of the levels of the risk factor under the specified attribute features includes:

[0014] For each of the risk factors, the weight of the risk factor is determined by a pre-trained decision tree, which includes multiple nodes. For each node, each node represents a risk factor and its attribute features. Each node corresponds to the weight of the risk factor corresponding to that node. For every two nodes, the connection represents the relationship between the two nodes.

[0015] For each of the risk factors, the level score corresponding to the level score of the risk factor under the specified attribute feature is determined by a first correspondence between each of the risk factors and the level score corresponding to each risk factor under different attribute features, wherein each attribute feature includes the specified attribute feature.

[0016] The beneficial effect of adopting the above-mentioned further scheme is that the weight of each risk factor, as well as the level score and value of each risk factor, can be determined more accurately through the pre-trained decision tree and the pre-established first correspondence.

[0017] Furthermore, for each of the aforementioned nodes, the weight of the node is determined in the following manner:

[0018] For each node, obtain the training sample corresponding to the node. The training sample includes multiple risk factors of the same type and the attribute features of each risk factor.

[0019] For each node, the Gini coefficient corresponding to the node is determined based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor.

[0020] For each node, the weight corresponding to the node is determined based on the Gini coefficient corresponding to the node.

[0021] The beneficial effect of adopting the above-mentioned further scheme is that the Gini coefficient can more accurately reflect the risk difference of each risk factor among all risk factors. Therefore, the weight of each risk factor can be accurately determined by the Gini coefficient.

[0022] Furthermore, the aforementioned first correspondence was determined in the following way:

[0023] For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor corresponding to the node under the specified attribute characteristics is determined. The level score represents the proportion of the number of risk factors with the same specified attribute characteristics in the training sample to the total number of samples in the training sample.

[0024] For each node, the total score of the risk factor corresponding to the node under the specified attribute feature is determined based on the level score corresponding to the node.

[0025] The first correspondence is established based on the risk factors corresponding to each node and the total score of each node.

[0026] The beneficial effect of adopting the above-mentioned further scheme is that if a risk factor has at least one attribute feature, then based on the attribute feature of the risk factor, the level score of each risk factor can be determined based on the training sample.

[0027] Furthermore, for each node, the determination of the Gini coefficient corresponding to the node based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor includes:

[0028] For each node, the Gini coefficient corresponding to the node is determined by a first formula based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, wherein the first formula is:

[0029]

[0030] Where D represents the risk factors of the same type corresponding to the nodes, Gini(D) represents the Gini coefficient, k represents the number of attribute characteristics of the risk factors, |D| represents the number of risk factors, and |C k | indicates the number of the k-th attribute feature.

[0031] The advantage of adopting the above-mentioned further scheme is that the Gini coefficient corresponding to each node can be accurately determined through the first formula mentioned above.

[0032] Furthermore, for each node, the above-mentioned determination of the level score of the risk factor corresponding to the node under the specified attribute characteristics based on the attribute characteristics of the risk factor corresponding to the node includes:

[0033] For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor under the specified attribute characteristics is determined by a second formula, wherein the second formula is:

[0034]

[0035] Among them, H ik T represents the level score corresponding to the k-th attribute feature of the i-th risk factor. ik S represents the number of risk factors in the training samples that have the k-th attribute feature. ik This represents the total number of samples in the training samples, and the k-th attribute feature is the specified attribute feature.

[0036] The beneficial effect of adopting the above-mentioned further scheme is that the level score corresponding to each risk factor under the specified attribute characteristics can be accurately determined through the above-mentioned second formula.

[0037] Furthermore, for each node, the above-mentioned determination of the horizontal total score corresponding to the node based on the horizontal score corresponding to the node includes:

[0038] For each node, the total horizontal score corresponding to the node is determined using a third formula based on the horizontal score of the node. The third formula is:

[0039]

[0040] Among them, f ik Let n represent the total score of the k-th attribute feature corresponding to the i-th risk factor. i This represents the number of attribute characteristics possessed by the i-th risk factor.

[0041] The advantage of adopting the above-mentioned further scheme is that the horizontal total score corresponding to each node can be accurately determined through the third formula mentioned above.

[0042] Secondly, in order to solve the above-mentioned technical problems, the present invention also provides a risk ranking device for circumferential welds of in-service pipelines, the device comprising:

[0043] The risk factor acquisition module is used to acquire multiple risk factors for welds to be prioritized.

[0044] The weight and level score determination module is used to obtain, for each risk factor, the weight corresponding to the risk factor and the level score corresponding to the risk factor under a specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor.

[0045] The comprehensive risk value determination module is used to determine the comprehensive risk value of each risk factor based on its weight and level score.

[0046] The risk ranking module is used to rank the risks of pipeline circumferential welds based on the comprehensive risk value of each of the aforementioned risk factors.

[0047] Thirdly, in order to solve the above-mentioned technical problems, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the in-service pipeline circumferential weld risk ranking method of the present application.

[0048] Fourthly, in order to solve the above-mentioned technical problems, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the in-service pipeline circumferential weld risk ranking method of the present application.

[0049] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below.

[0051] Figure 1 A flowchart illustrating a risk prioritization method for circumferential welds in in-service pipelines, provided as an embodiment of the present invention;

[0052] Figure 2 A schematic flowchart illustrating another method for risk ranking of circumferential welds in in-service pipelines, provided as an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a risk prioritization device for circumferential welds of in-service pipelines provided in one embodiment of the present invention;

[0054] Figure 4This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0055] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0056] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0057] The solution provided by this invention can be applied to any application scenario that requires prioritizing the risks of circumferential welds in in-service pipelines. The solution provided by this invention can be executed by any electronic device, such as a user's terminal device, including at least one of the following: smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, smart TV, or smart in-vehicle device.

[0058] This invention provides a possible implementation, such as... Figure 1 As shown, a flowchart of a risk prioritization method for circumferential welds in in-service pipelines is provided. This method can be executed by any electronic device, such as a terminal device, or jointly executed by a terminal device and a server. For ease of description, the method provided in this embodiment will be described below using a terminal device as the execution subject as an example. Figure 1 The flowchart shown indicates that the method may include the following steps:

[0059] Step S110: Obtain multiple risk factors for the weld joints to be prioritized.

[0060] Step S120: For each risk factor, obtain the weight corresponding to the risk factor and the level score corresponding to the risk factor under the specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor.

[0061] Step S130: For each risk factor, determine the comprehensive risk value of the risk factor based on its weight and level score.

[0062] Step S140: Based on the comprehensive risk value of each of the aforementioned risk factors, the risks of the pipeline circumferential weld are ranked.

[0063] This application's solution considers multiple risk factors affecting pipeline weld risks, as well as the risk level of each risk factor relative to other risk factors, and the risk level of each risk factor's specified attribute characteristics relative to other attribute characteristics of that risk factor. It determines the comprehensive risk value of each risk factor from multiple perspectives, and finally, based on the comprehensive risk values ​​of each risk factor, it ranks the risks of pipeline circumferential welds. This solution can guide the current domestic circumferential weld inspection, effectively reducing excavation volume and improving excavation accuracy.

[0064] The following specific embodiments further illustrate the solution of the present invention. In this embodiment, a risk ranking method for circumferential welds of in-service pipelines may include the following steps:

[0065] Step S110: Obtain multiple risk factors for the weld joints to be prioritized.

[0066] Among them, the above-mentioned multiple risk factors refer to factors that affect the risk of the weld joint, including but not limited to at least two of the following: inspection and classification, conveying medium, design pressure, steel grade, pipe diameter, welding method, wall thickness difference, weld joint process (gold joint), weld joint type (joint / damped joint), whether it is a repair joint, whether it is a bend joint, whether it is a variable wall thickness joint, whether it is a weld joint with questionable film, construction year, construction season, construction unit, testing unit, supervision unit, terrain, burial depth, whether the weld is abnormal, the severity of the weld abnormality, the circumferential length of the weld abnormality, distance from the joint weld joint, and distance from the bend joint.

[0067] Each risk factor corresponds to at least one attribute feature. An attribute feature refers to the characteristic that reflects the risk factor's attributes. For example, "investigation and classification" refers to the reasons for excavation, including questionable radiographs, abnormal internal inspections, or random checks. "Golden joint" refers to the welding process of the weld joint; therefore, the attribute features corresponding to "golden joint" include "is golden joint" (meaning the welding process of the weld joint is golden joint) and "is not golden joint" (meaning the welding process of the weld joint is not golden joint). Similarly, the above "yes" and "no" statements all include two different results: one corresponding to "yes" and one corresponding to "no." The specific attribute features also correspond to two different results. For example, "is it a bend in the pipe?" corresponds to "is a bend in the pipe?" and "is not a bend in the pipe?". Furthermore, for the description of risk factors using "yes" and "no" statements, each risk factor can also correspond to at least one attribute feature. For example, "construction unit" can correspond to two attribute features for either unit A or unit B.

[0068] The aforementioned risk factors can be obtained through manual statistical methods.

[0069] Step S120: For each risk factor, obtain the weight corresponding to the risk factor and the level score corresponding to the risk factor under the specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor.

[0070] It is understandable that for each risk factor, the corresponding attribute characteristics are different, and therefore the specified attribute characteristics for each risk factor are also different. For a given risk factor, the specified attribute characteristic refers to any one of at least one of the attribute characteristics corresponding to that risk factor.

[0071] Optionally, for each of the risk factors, obtaining the weight corresponding to the risk factor and the level score of the risk factor under the specified attribute feature includes:

[0072] For each risk factor, the weight of the risk factor is determined by a pre-trained decision tree, which includes multiple nodes. For each node, each node represents a risk factor and its attribute features, and each node corresponds to the weight of the risk factor corresponding to that node. For the connection between any two nodes, the connection represents the relationship between the two nodes. As an example, the two nodes are node A and node B. The connection between node A and node B represents the risk factor corresponding to node B under the risk factor corresponding to node A. That is, the connection represents the hierarchical relationship between the risk factor corresponding to node A and the risk factor corresponding to node B.

[0073] For each of the risk factors, the level score corresponding to the level score of the risk factor under the specified attribute feature is determined by a first correspondence between each of the risk factors and the level score corresponding to each risk factor under different attribute features, wherein each attribute feature includes the specified attribute feature.

[0074] Optionally, the pre-trained decision tree described above can be constructed in the following ways:

[0075] (1) The data after rule-based processing (multiple preprocessed risk factors) is used to construct a training sample set and a test sample set. The ratio of the number of samples in the training sample set to the number of samples in the test sample set can be 4:1. The training sample set includes multiple risk factors corresponding to different pipe welds, as well as the attribute features corresponding to each risk factor. Among them, the data of each risk factor obtained initially is disordered and needs to be preprocessed to rule out the disordered and repetitive content in the data.

[0076] (2) Based on the above training sample set, the nodes are divided. According to the sample constraints corresponding to each node at each level, the dataset D corresponding to each node is determined. The sample constraints can be set based on the attribute characteristics of each risk factor. For example, in the training sample set, the attribute characteristic of a certain risk factor is A, and the number of samples with the attribute characteristic of a certain risk factor is A is not less than the set number. In this way, according to the set sample constraints, each sample in the training sample set can be divided into the corresponding node.

[0077] (3) The above sample constraints may also include conditions determined based on the Gini coefficient of each sample. After satisfying the conditions in (2), further processing can be performed based on the Gini coefficient. If the Gini coefficient is less than the threshold, the decision tree subtree is returned and the current node stops recursion. That is, if the Gini coefficient is less than the threshold, it means that the risk factor corresponding to the Gini coefficient has a small impact on the overall risk and can be ignored. Therefore, the sample division for the current node can be stopped. It should be noted that a node can correspond to multiple identical risk factors in the training sample set.

[0078] (4) For a current node, based on the attribute characteristics of the risk factors corresponding to the current node, the current node can be split into two leaf nodes, and the dataset corresponding to the current node can be divided into two subsets. The specific implementation process is as follows:

[0079] Calculate the Gini coefficient of each feature (attribute feature) of the current node with respect to the dataset D. Among the calculated Gini coefficients of each feature with respect to the dataset D, select feature A and its corresponding feature value a, which have the smallest Gini coefficients. These are then designated as the optimal feature and optimal feature value, respectively. Based on these optimal features and optimal feature values, the dataset corresponding to the current node can be divided into two parts, D1 and D2. Simultaneously, left and right nodes are established for the current node. The dataset D of the left node is designated as D1, and the dataset D of the right node is designated as D2. In other words, the dataset corresponding to the current node is divided into two leaf nodes (left and right nodes), corresponding to the subsets D1 and D2. The current node and its two leaf nodes can be connected by lines.

[0080] After obtaining the Gini coefficient for each sample, the weight corresponding to each sample, i.e., the weight corresponding to each risk factor, can be determined based on the Gini coefficient. The specific implementation process is as follows:

[0081] For each node, obtain the training sample corresponding to the node. The training sample includes multiple risk factors of the same type and the attribute features of each risk factor.

[0082] For each node, the Gini coefficient corresponding to the node is determined based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor.

[0083] For each node, the weight corresponding to the node is determined based on the Gini coefficient corresponding to the node.

[0084] Optionally, for each node, determining the Gini coefficient corresponding to the node based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor includes:

[0085] For each node, the Gini coefficient corresponding to the node is determined by a first formula based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, wherein the first formula is:

[0086]

[0087] Where D represents the risk factors of the same type corresponding to the nodes, Gini(D) represents the Gini coefficient, k represents the number of attribute characteristics of the risk factors, |D| represents the number of each risk factor, and |C k | indicates the number of the k-th attribute feature.

[0088] For risk factors D of the same type, numbered |D|, and based on a certain value 'a' (a specified attribute feature) of feature A, D is divided into two datasets, |D1| and |D2|. Then, under the condition of feature A, the expression for the Gini coefficient of risk factors D of the same type is:

[0089]

[0090] In the scheme of this application, for each node, the Gini coefficient corresponding to that node can be either Gini(D) or Gini(D, A).

[0091] After determining the Gini coefficient for each node, for each node, determining the weight corresponding to the node based on the Gini coefficient can include:

[0092] Suppose a node corresponds to K risk factors, and the probability of the k-th risk factor is p. k Gini coefficient, or weighting coefficient, is the probability that the number of samples corresponding to the Kth risk factor is relative to the total number of samples at that node.

[0093]

[0094] Where Gini(p) represents the weight of a node, and p represents the Gini coefficient of each node.

[0095] In the scheme of this application, for each node, the weight corresponding to that node can be obtained based on the pre-trained decision tree.

[0096] (5) Through the above node partitioning method, multiple nodes can be obtained. Each node corresponds to a dataset, which includes at least one risk factor that satisfies the sample constraint conditions, as well as the attribute characteristics of each risk factor. Each node also corresponds to a weight. Through the above multiple nodes, the weights corresponding to each node and the connections between each node, a decision tree can be established.

[0097] (6) Use the test sample set to verify the performance of the decision tree classification model. If the performance verification is successful, the decision tree classification model is considered to have been trained. Otherwise, proceed to step 3.

[0098] Optionally, the above first correspondence is determined in the following way:

[0099] For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor corresponding to the node under the specified attribute characteristics is determined. The level score represents the proportion of the number of risk factors with the same specified attribute characteristics in the training sample to the total number of samples in the training sample.

[0100] For each node, the total score of the risk factor corresponding to the node under the specified attribute feature is determined based on the level score corresponding to the node.

[0101] The first correspondence is established based on the risk factors corresponding to each node and the total score of each node.

[0102] Optionally, for a given risk factor, its attribute characteristics will affect the score f. i The size, that is

[0103] f ik =f(H ik )

[0104] Optionally, for each node, the above-mentioned determination of the level score of the risk factor corresponding to the node under the specified attribute characteristics based on the attribute characteristics of the risk factor corresponding to the node includes:

[0105] For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor under the specified attribute characteristics is determined by a second formula, wherein the second formula is:

[0106]

[0107] Among them, H ik T represents the level score corresponding to the k-th attribute feature of the i-th risk factor, with a value range of 0 to 1. ik S represents the number of risk factors in the training samples that have the k-th attribute feature. ik This represents the total number of samples in the training samples, and the k-th attribute feature is the specified attribute feature.

[0108] Optionally, for each node, determining the total horizontal score corresponding to the node based on the horizontal score corresponding to the node includes:

[0109] For each node, the total horizontal score corresponding to the node is determined using a third formula based on the horizontal score of the node. The third formula is:

[0110]

[0111] Among them, f ik Let n represent the total score of the k-th attribute feature corresponding to the i-th risk factor. i This represents the number of attribute characteristics possessed by the i-th risk factor.

[0112] Step S130: For each risk factor, determine the comprehensive risk value of the risk factor based on its weight and level score.

[0113] After determining the weight and level score of each risk factor, the comprehensive risk value of the risk factor can be determined based on the weight and level score of the risk factor. Specifically, for each risk factor, the weight and level score of the risk factor are multiplied together, and the product is used as the comprehensive risk value of the risk factor.

[0114] Optionally, the risk value corresponding to the weld joint to be ranked can be determined based on the comprehensive risk value corresponding to each risk factor. The risk value corresponding to the weld joint to be ranked can be expressed by the following formula:

[0115]

[0116] Where F is the risk value, α i f represents the weight corresponding to the i-th risk factor. i Let M be the level score corresponding to the i-th risk factor, and M represent the total number of risk factors corresponding to the weld joints to be ranked.

[0117] Step S140: Based on the comprehensive risk value of each of the aforementioned risk factors, the risks of the pipeline circumferential weld are ranked.

[0118] Specifically, the combined risk values ​​of the aforementioned risk factors can be ranked from highest to lowest, with higher scores indicating greater risk.

[0119] To better illustrate and understand the principle of the method provided by this invention, the following description uses an optional specific embodiment to illustrate the solution of this invention. It should be noted that the specific implementation of each step in this specific embodiment should not be construed as a limitation of the solution of this invention. Other implementations that can be conceived by those skilled in the art based on the principle of the solution provided by this invention should also be considered within the scope of protection of this invention.

[0120] In this example, see Figure 2 First, based on the method described above, a decision tree needs to be trained and the initial correspondence established. This begins with data collection and processing, resulting in multiple samples, including various risk factors corresponding to different pipe welds, and the attribute features of each risk factor. This data is then used to construct training and testing sets. Next, a decision tree model is trained based on these sets, resulting in a pre-trained decision tree analysis model. Each node in this decision tree corresponds to the same type of risk factor, and the weight of each node can be determined through the decision tree. Based on each sample in the training set, the initial correspondence between each risk factor and each level's total score can be determined using the method described above.

[0121] After obtaining the first correspondence and decision tree, in practical applications, for weld joints to be ranked by risk, data collection is first performed, i.e., multiple risk factors corresponding to the weld joints to be ranked by risk are obtained. Then, the data of multiple risk factors is preprocessed, and the processed data is input into the trained decision tree analysis model to determine the weight corresponding to each risk factor, i.e., the weight of each factor is output. At the same time, based on the first correspondence, the level score corresponding to each risk factor is determined, i.e., the level score is calculated manually. Then, based on the weight and level score of each risk factor, the comprehensive risk value of each risk factor is determined (weld joint risk score calculation). The risk value corresponding to the weld joints to be ranked by risk can also be determined based on the comprehensive risk value of each risk factor.

[0122] This invention can be used for risk ranking of weld joints in in-service gas pipelines, and can guide the current domestic inspection of circumferential welds, effectively reducing the amount of excavation and improving the excavation accuracy.

[0123] Based on and Figure 1 Based on the same principle as the method shown, this embodiment of the invention also provides a risk ranking device 20 for in-service pipeline circumferential welds, such as... Figure 3As shown, the in-service pipeline circumferential weld risk ranking device 20 may include a risk factor acquisition module 210, a weight and level total score determination module 220, a comprehensive risk value determination module 230, and a risk ranking module 240, wherein:

[0124] Risk factor acquisition module 210 is used to acquire multiple risk factors of weld joints to be ranked by risk.

[0125] The weight and level score determination module 220 is used to obtain, for each risk factor, the weight corresponding to the risk factor and the level score corresponding to the risk factor under a specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor.

[0126] The comprehensive risk value determination module 230 is used to determine the comprehensive risk value of each risk factor based on its weight and level score.

[0127] The risk ranking module 240 is used to rank the risks of pipeline circumferential welds based on the comprehensive risk value of each of the aforementioned risk factors.

[0128] Optionally, for each of the aforementioned risk factors, the weight and level score determination module 220, when obtaining the weight corresponding to the risk factor and the level score corresponding to the risk factor under the specified attribute feature, is specifically used for:

[0129] For each of the risk factors, the weight of the risk factor is determined by a pre-trained decision tree, which includes multiple nodes. For each node, each node represents a risk factor and its attribute features. Each node corresponds to the weight of the risk factor corresponding to that node. For every two nodes, the connection represents the relationship between the two nodes.

[0130] For each of the risk factors, the level score corresponding to the level score of the risk factor under the specified attribute feature is determined by a first correspondence between each of the risk factors and the level score corresponding to each risk factor under different attribute features, wherein each attribute feature includes the specified attribute feature.

[0131] Optionally, for each node, the weight of the node is determined by the following weight module: the weight module is used to obtain training samples corresponding to each node, wherein the training samples include multiple risk factors of the same type and attribute features of each risk factor; for each node, determine the Gini coefficient corresponding to the node based on each risk factor and the attribute features corresponding to each risk factor; and for each node, determine the weight corresponding to the node based on the Gini coefficient corresponding to the node.

[0132] Optionally, the aforementioned first correspondence is determined by the following correspondence determination module, which is used to, for each node, determine the level score of the risk factor corresponding to the node under the specified attribute feature based on the attribute feature of the risk factor corresponding to the node, wherein the level score represents the proportion of the number of risk factors with the same specified attribute feature in the training sample to the total number of samples in the training sample; for each node, determine the total level score of the risk factor corresponding to the node under the specified attribute feature based on the level score of the node; and establish the first correspondence based on the risk factors corresponding to each node and the total level score of each node.

[0133] Optionally, for each node, when the weighting module determines the Gini coefficient corresponding to the node based on the various risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, it is specifically used for:

[0134] For each node, the Gini coefficient corresponding to the node is determined by a first formula based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, wherein the first formula is:

[0135]

[0136] Where D represents the risk factors of the same type corresponding to the nodes, Gini(D) represents the Gini coefficient, k represents the number of attribute characteristics of the risk factors, |D| represents the number of risk factors, and |C k | indicates the number of the k-th attribute feature.

[0137] Optionally, for each node, when the above-mentioned correspondence determination module determines the level score of the risk factor corresponding to the node under the specified attribute characteristics based on the attribute characteristics of the risk factor corresponding to the node, it is specifically used for:

[0138] For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor under the specified attribute characteristics is determined by a second formula, wherein the second formula is:

[0139]

[0140] Among them, H ik T represents the level score corresponding to the k-th attribute feature of the i-th risk factor. ik S represents the number of risk factors in the training samples that have the k-th attribute feature. ik This represents the total number of samples in the training samples, and the k-th attribute feature is the specified attribute feature.

[0141] Optionally, for each node, when the above-mentioned correspondence determination module determines the total horizontal score corresponding to the node based on the horizontal score corresponding to the node, it is specifically used for:

[0142] For each node, the total horizontal score corresponding to the node is determined using a third formula based on the horizontal score of the node. The third formula is:

[0143]

[0144] Among them, f ik Let n represent the total score of the k-th attribute feature corresponding to the i-th risk factor. i This represents the number of attribute characteristics possessed by the i-th risk factor.

[0145] The in-service pipeline circumferential weld risk ranking device of this invention can execute the in-service pipeline circumferential weld risk ranking method provided in this invention. The implementation principle is similar. The actions performed by each module and unit in the in-service pipeline circumferential weld risk ranking device of each embodiment of this invention correspond to the steps in the in-service pipeline circumferential weld risk ranking method of each embodiment of this invention. For a detailed functional description of each module of the in-service pipeline circumferential weld risk ranking device, please refer to the description of the corresponding in-service pipeline circumferential weld risk ranking method shown above, which will not be repeated here.

[0146] The aforementioned risk ranking device for circumferential welds of in-service pipelines can be a computer program (including program code) running on a computer device, such as an application software; the device can be used to execute the corresponding steps in the method provided in the embodiments of the present invention.

[0147] In some embodiments, the in-service pipeline circumferential weld risk ranking device provided by the present invention can be implemented in a combination of hardware and software. As an example, the in-service pipeline circumferential weld risk ranking device provided by the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the in-service pipeline circumferential weld risk ranking method provided by the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0148] In other embodiments, the risk ranking device for circumferential welds of in-service pipelines provided in this invention can be implemented in software. Figure 3 A risk ranking device for in-service pipeline circumferential welds stored in a memory is shown. It can be software in the form of programs and plug-ins, and includes a series of modules, including a risk factor acquisition module 210, a weight and level total score determination module 220, a comprehensive risk value determination module 230, and a risk ranking module 240, for implementing a risk ranking method for in-service pipeline circumferential welds provided in this embodiment of the invention.

[0149] The modules described in the embodiments of the present invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0150] Based on the same principles as the methods shown in the embodiments of the present invention, the embodiments of the present invention also provide an electronic device, which may include, but is not limited to: a processor and a memory; the memory for storing computer programs; and the processor for executing the methods shown in any embodiment of the present invention by invoking the computer programs.

[0151] In one alternative embodiment, an electronic device is provided, such as Figure 4 As shown, Figure 4The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0152] 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 the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0153] Bus 4002 may include a pathway for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0154] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0155] The memory 4003 stores the application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0156] Among these, electronic devices can also be terminal devices. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0157] This invention provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0158] According to another aspect of the present invention, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the in-service pipeline circumferential weld risk ranking method provided in the various embodiments described above.

[0159] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0160] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0161] The computer-readable storage medium provided in this invention can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a 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, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0162] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0163] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

Claims

1. A method for risk ranking of circumferential welds in in-service pipelines, characterized in that, Includes the following steps: Obtain multiple risk factors for welds to be prioritized; For each risk factor, obtain the weight corresponding to the risk factor and the level score corresponding to the risk factor under the specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor. For each of the aforementioned risk factors, a comprehensive risk value for that risk factor is determined based on its weight and level score. Based on the comprehensive risk value of each of the aforementioned risk factors, the risks of pipeline circumferential welds are ranked. For each of the risk factors, obtaining the weight corresponding to the risk factor and the sum of the levels of the risk factor under the specified attribute features includes: For each of the risk factors, the weight of the risk factor is determined by a pre-trained decision tree, which includes multiple nodes. For each node, each node represents a risk factor and its attribute features. Each node corresponds to the weight of the risk factor corresponding to that node. For every two nodes, the connection represents the relationship between the two nodes. For each of the risk factors, the level score corresponding to the level score of the risk factor under the specified attribute feature is determined by a first correspondence between each of the risk factors and the level score corresponding to each risk factor under different attribute features, wherein each attribute feature includes the specified attribute feature. The pre-trained decision tree is constructed in the following way: S1, constructs a training sample set and a test sample set from the preprocessed multiple risk factors; S2, divide the nodes according to the training sample set, and determine the dataset D corresponding to each node according to the pre-set sample constraints for each node at each level. The sample constraints can be set based on the attribute characteristics of each risk factor. S3, the sample constraint also includes the condition determined based on the Gini coefficient of each sample. After the sample constraint in S2 is satisfied, for each node, processing is performed based on the Gini coefficient. If the Gini coefficient is less than the threshold, the decision tree subtree is returned and the node stops recursion. S4. For each node, based on the attribute characteristics of the risk factors corresponding to that node, calculate the Gini coefficient of each attribute characteristic of each existing risk factor of that node with respect to dataset D; among the calculated Gini coefficients of each attribute characteristic of each risk factor with respect to dataset D, select the risk factor A with the smallest Gini coefficient and the corresponding attribute characteristic a, and take the risk factor A with the smallest Gini coefficient and the corresponding attribute characteristic a as the optimal feature and the optimal feature value, respectively. According to the optimal feature and the optimal feature value, divide the dataset corresponding to that node into two parts D1 and D2, and at the same time establish the left and right nodes of that node as the two leaf nodes of that node, and there are lines connecting that node and the two corresponding leaf nodes. Based on the risk factors corresponding to the node and the attribute characteristics of each risk factor, the Gini coefficient corresponding to the node is determined; assuming that the node corresponds to K risk factors, and the probability of the Kth risk factor is... Gini coefficient, or weighting coefficient, is the probability that the number of samples corresponding to the Kth risk factor is relative to the total number of samples at that node. Wherein, Gini(p) represents the weight corresponding to the node; S5. Through the node partitioning method described above, multiple nodes are obtained. Each node corresponds to a dataset, which includes at least one risk factor that satisfies the sample constraints, as well as the attribute features of each risk factor. Each node also corresponds to a weight. A decision tree is established through the multiple nodes, the weights corresponding to each node, and the connections between the nodes.

2. The method according to claim 1, characterized in that, The first correspondence was determined in the following way: For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor corresponding to the node under the specified attribute characteristics is determined. The level score represents the proportion of the number of risk factors with the same specified attribute characteristics in the training sample to the total number of samples in the training sample. For each node, the total score of the risk factor corresponding to the node under the specified attribute feature is determined based on the level score corresponding to the node. The first correspondence is established based on the risk factors corresponding to each node and the total score of each node.

3. The method according to claim 1, characterized in that, For each node, the Gini coefficient corresponding to the node is determined based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, including: For each node, the Gini coefficient corresponding to the node is determined by a first formula based on the risk factors corresponding to the node and the attribute characteristics corresponding to each risk factor, wherein the first formula is: Where D represents the risk factors of the same type corresponding to the nodes, Gini(D) represents the Gini coefficient, and k represents the number of attribute characteristics of the risk factors. Indicates the number of risk factors. This represents the number of the k-th attribute feature.

4. The method according to claim 2, characterized in that, For each node, determining the level score of the risk factor corresponding to the node under a specified attribute feature based on the attribute features of the risk factor corresponding to the node includes: For each node, based on the attribute characteristics of the risk factor corresponding to the node, the level score of the risk factor under the specified attribute characteristics is determined by a second formula, wherein the second formula is: Among them, H ik T represents the level score corresponding to the k-th attribute feature of the i-th risk factor. ik S represents the number of risk factors in the training samples that have the k-th attribute feature. ik This represents the total number of samples in the training samples, and the k-th attribute feature is the specified attribute feature.

5. The method according to claim 4, characterized in that, For each node, determining the total horizontal score corresponding to the node based on the horizontal score corresponding to the node includes: For each node, the total horizontal score corresponding to the node is determined using a third formula based on the horizontal score of the node. The third formula is: Among them, f ik Let n represent the total score of the k-th attribute feature corresponding to the i-th risk factor. i This represents the number of attribute characteristics possessed by the i-th risk factor.

6. A risk ranking device for circumferential welds of in-service pipelines, characterized in that, The device for the risk ranking method of circumferential welds of in-service pipelines according to claim 1 includes: The risk factor acquisition module is used to acquire multiple risk factors for welds to be prioritized. The weight and level score determination module is used to obtain, for each risk factor, the weight corresponding to the risk factor and the level score corresponding to the risk factor under a specified attribute feature. The weight represents the risk level of the risk factor relative to each other risk factor, and the level score represents the risk level of the specified attribute feature of the risk factor relative to each other attribute feature of the risk factor. The comprehensive risk value determination module is used to determine the comprehensive risk value of each risk factor based on its weight and level score. The risk ranking module is used to rank the risks of pipeline circumferential welds based on the comprehensive risk value of each of the aforementioned risk factors.

7. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1-5.

Images