CN116739188B - Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect - Google Patents
Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect Download PDFInfo
- Publication number
- CN116739188B CN116739188B CN202311007712.1A CN202311007712A CN116739188B CN 116739188 B CN116739188 B CN 116739188B CN 202311007712 A CN202311007712 A CN 202311007712A CN 116739188 B CN116739188 B CN 116739188B
- Authority
- CN
- China
- Prior art keywords
- pipeline
- pipe section
- vulnerability
- index
- indexes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000000694 effects Effects 0.000 title claims description 6
- 238000012216 screening Methods 0.000 claims abstract description 32
- 239000013598 vector Substances 0.000 claims description 35
- 210000002569 neuron Anatomy 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 21
- 239000011435 rock Substances 0.000 claims description 10
- 239000002689 soil Substances 0.000 claims description 10
- 238000004590 computer program Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000009933 burial Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000009966 trimming Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/24—Earth materials
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Chemical & Material Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Food Science & Technology (AREA)
- Environmental & Geological Engineering (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geology (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Medicinal Chemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a device for early warning the vulnerability of an oil and gas pipeline under the action of geological disasters, which can improve the accuracy of pipeline vulnerability prediction by constructing corresponding vulnerability prediction models for different geological conditions and geological disaster types; meanwhile, indexes with higher importance are obtained by screening after importance analysis is carried out on disaster body indexes and pipeline indexes, and are used as influence indexes, so that indexes which are more effective for predicting pipeline vulnerability can be screened more objectively and more accurately; the index values of the pipe section to be tested, corresponding to the influence indexes of the vulnerability prediction model, are input into the vulnerability prediction model to obtain the vulnerability grade of the pipe section to be tested and the damage description information of the reference pipe section, which are output by the vulnerability prediction model, so that a user can more intuitively know the damage condition of the pipe section to be tested, and in addition, the damage description information of the reference pipe section can objectively and accurately reflect the damage condition of the pipe section to be tested under the action of geological disasters.
Description
Technical Field
The invention relates to the technical field of pipeline prediction, in particular to an early warning method and device for vulnerability of an oil and gas pipeline under the action of geological disasters.
Background
Pipeline transportation is the most common transportation mode at present, however, oil and gas pipelines are subject to various risks including third party damage to the pipelines, pipeline corrosion, and the influence of geological disasters while developing at high speed. Among them, geological disasters are a very important factor threatening the safety of pipelines, which can cause serious damage to the pipelines themselves. However, after a geological disaster occurs, there are many inconveniences due to the long oil and gas pipeline and the influence of the geological disaster, and immediate measurement and evaluation of damage thereof is extremely difficult. Therefore, the early warning is necessary to be carried out on the possible damage condition of the pipeline under the influence of the current geological disaster, and proper post-disaster treatment measures are timely taken, so that serious secondary damage caused by the damage of the pipeline due to the geological disaster is avoided.
Currently, in order to evaluate the damage condition of a pipeline (i.e., pipeline vulnerability) under the influence of geological disasters, the following methods are generally adopted: the method comprises the steps of manually constructing an index system, wherein most of indexes are selected to compare the attribute of a pipeline and the position relation between the pipeline and a geological disaster, calculating objective weights of vulnerability indexes by using an entropy weight method after selecting proper evaluation indexes, determining the objective weights according to the difference of the indexes by using the basic thought of the entropy weights, and then carrying out weighted summation on index values of the indexes by using the objective weights to obtain the vulnerability score of the pipeline. However, in the method, the accuracy of the vulnerability score obtained by calculation is directly affected by the construction of the index system, but the artificial construction of the index system has the defect of dependence on subjective experience, and under the condition of geological disasters and the complex condition of the attribute diversification of the pipelines, the damage degree of different pipelines under the influence of various geological disasters is difficult to accurately reflect by the artificially constructed index system. In addition, the current pipeline vulnerability assessment method can only output a vulnerability score or a vulnerability grade calculated according to a self-built index system, however, the vulnerability score or the vulnerability grade has strong correlation with the index system, has abstract description capability and limitation on pipeline damage, and cannot objectively and intuitively describe possible damage conditions of the pipeline.
Disclosure of Invention
The invention provides an early warning method and device for vulnerability of an oil and gas pipeline under the action of geological disasters, which are used for solving the defects that an index system is not accurately established manually in the prior art, the description capacity of pipeline damage is abstract and has limitations, and the possible damage condition of the pipeline cannot be objectively and intuitively described.
The invention provides an early warning method for vulnerability of an oil and gas pipeline under the action of geological disasters, which comprises the following steps:
dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
According to the oil and gas pipeline vulnerability early warning method under the action of geological disasters, each influence index of the vulnerability prediction model corresponding to any pipe section to be tested is determined based on the following steps:
iterative training is carried out on the initial feature selection model based on index values of a plurality of sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines; the type of geological disaster faced by the sample pipeline and the geological conditions of the sample pipeline are the same as those of any pipe section to be tested; the characteristic selection model and the vulnerability prediction model corresponding to any pipe section to be tested have the same structure except that the number of neurons of an input layer is different, and the neurons of the input layer in the characteristic selection model correspond to disaster indexes and pipeline indexes one by one; neurons of an input layer in the vulnerability prediction model correspond to the influence indexes one by one;
Determining a first importance degree of the disaster index and the pipeline index based on weights corresponding to each neuron of the input layer in the feature selection model after iterative training;
and screening influence indexes from the disaster body indexes and the pipeline indexes based on the first importance degrees of the disaster body indexes and the pipeline indexes.
According to the method for early warning the vulnerability of the oil and gas pipeline under the action of geological disasters, provided by the invention, the impact indexes are screened from the disaster indexes and the pipeline indexes based on the first importance degree of the disaster indexes and the pipeline indexes, and the method specifically comprises the following steps:
predicting the index values of the disaster indexes and the pipeline indexes corresponding to the sample pipelines by utilizing the feature selection model after the iterative training, so as to obtain a first probability of the vulnerability grades corresponding to the sample pipelines;
the disaster body index and/or the pipeline index with the first importance degree being lower than a first preset threshold value are selected as candidate indexes, and the disaster body index and/or the pipeline index with the first importance degree being greater than or equal to the first preset threshold value are selected as influence indexes;
carrying out randomization treatment on index values of each sample pipeline corresponding to any candidate index, and then, utilizing the feature selection model after iterative training, predicting the index values of each sample pipeline corresponding to each disaster body index and pipeline index to obtain a second probability of each sample pipeline corresponding to each vulnerability grade;
Determining a second importance degree of any candidate index based on a difference between the first probability and the second probability of each sample pipe corresponding to each vulnerability class;
and screening the candidate indexes for influencing indexes based on the second importance degree of the candidate indexes.
According to the method for early warning the vulnerability of the oil and gas pipeline under the action of geological disasters, the second important degree of any candidate index is determined based on the difference between the first probability and the second probability of each sample pipeline corresponding to each vulnerability level, and the method specifically comprises the following steps:
summing the differences between the first probability and the second probability of each vulnerability grade corresponding to any sample pipeline to obtain the corresponding characteristic loss of any sample pipeline;
and summing the characteristic losses corresponding to the sample pipelines to obtain a second importance degree of any candidate index.
According to the method for early warning the vulnerability of the oil and gas pipeline under the action of geological disasters, which is provided by the invention, the influence indexes are screened from the candidate indexes based on the second important degree of the candidate indexes, and the method specifically comprises the following steps:
determining the comprehensive importance degree of any candidate index based on the first importance degree and the second importance degree of the any candidate index aiming at the any candidate index;
And if the comprehensive importance degree of any candidate index is greater than a second preset threshold value, determining the any candidate index as an influence index.
According to the early warning method for the vulnerability of the oil and gas pipeline under the action of geological disasters, the reference pipe section is determined based on the following steps:
acquiring the latest index values of the corresponding influence indexes of the known pipe sections which are of the same type, are in the same geological condition as any pipe section to be tested and have known damage conditions before the occurrence of the geological disaster;
inputting the latest index values of the known pipe sections corresponding to the influence indexes into the vulnerability prediction model respectively to obtain vulnerability levels of the known pipe sections output by the vulnerability prediction model;
selecting a known pipe section with the vulnerability grade identical to that of any pipe section to be tested as a candidate pipe section;
determining a pipeline state vector of the candidate pipe section based on the latest index value of the pipeline influence index in the influence indexes corresponding to the candidate pipe section, and determining the pipeline state vector of any pipe section to be tested based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section to be tested; the pipeline state vector characterizes the pipeline state of the corresponding pipeline section;
And screening the reference pipe section from the candidate pipe sections based on the distance between the pipeline state vector of any pipe section to be tested and the pipeline state vector of each candidate pipe section.
According to the early warning method for the vulnerability of the oil and gas pipeline under the action of geological disasters, the pipeline state vector of the candidate pipeline section or any pipeline section to be tested is determined based on the following steps:
respectively normalizing index values of the pipeline influence indexes in the candidate pipeline sections or any pipeline section to be tested corresponding to the influence indexes to obtain normalized index values of the pipeline influence indexes corresponding to the corresponding pipeline sections;
determining the weight of each pipeline influence index based on the weight of each neuron of the input layer in the vulnerability prediction model;
and weighting and combining the normalized index values of the candidate pipe section or any pipe section to be tested corresponding to the pipeline influence indexes based on the weight of the pipeline influence indexes to obtain the pipeline state vector of the candidate pipe section or any pipe section to be tested.
According to the oil and gas pipeline vulnerability early warning method under the geological disaster effect, the vulnerability prediction model is trained based on index values of a plurality of sample pipelines corresponding to all influence indexes and vulnerability grade labels of the corresponding sample pipelines.
The invention provides an early warning method for vulnerability of an oil and gas pipeline under the action of geological disasters, which further comprises the following steps:
and after the actual damage condition of each pipe section to be detected in the target pipeline is obtained, fine-tuning the vulnerability prediction model based on the actual damage condition of each pipe section to be detected.
The invention also provides an early warning device for vulnerability of the oil and gas pipeline under the action of geological disasters, which comprises:
the pipe section dividing unit is used for dividing the target pipeline into a plurality of independent pipe sections to be measured; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
the vulnerability prediction unit is used for determining a vulnerability prediction model corresponding to any pipe section to be detected based on the current geological disaster type and geological conditions of the any pipe section to be detected, inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be detected into the vulnerability prediction model, and obtaining vulnerability grade of the any pipe section to be detected and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
The invention also provides electronic equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the early warning method of the vulnerability of the oil and gas pipeline under the action of any geological disaster when executing the program.
The invention also provides a non-transitory computer readable storage medium, on which is stored a computer program which, when executed by a processor, implements a method for pre-warning of vulnerability of an oil and gas pipeline under the action of any one of the geological disasters described above.
The invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the early warning method of the vulnerability of the oil and gas pipeline under the action of any geological disaster when being executed by a processor.
According to the method and the device for early warning the vulnerability of the oil and gas pipeline under the action of the geological disaster, the pipeline vulnerability prediction accuracy can be improved by constructing the corresponding vulnerability prediction model for different geological conditions and geological disaster type combinations and learning the prediction mode of the pipeline damage under the combination of the geological conditions and the geological disaster type from the pipeline data of the sample pipeline under the combination of the corresponding geological conditions and the geological disaster type; meanwhile, index values corresponding to each disaster body index and pipeline index and vulnerability grade labels of corresponding sample pipelines are based on the sample pipelines, indexes with higher importance are obtained by screening after importance analysis is carried out on the disaster body index and the pipeline index, and are used as influence indexes, so that indexes which are more effective in predicting pipeline vulnerability can be screened more objectively and more accurately, and influence of subjective experience restriction when indexes are selected manually is avoided; after the trained vulnerability prediction model corresponding to any pipe section to be detected is determined, the index value of each influence index of the pipe section to be detected corresponding to the vulnerability prediction model is input into the vulnerability prediction model, so that the vulnerability grade of the pipe section to be detected and the damage description information of the reference pipe section output by the vulnerability prediction model are obtained, a user can more intuitively know the damage condition of the pipe section to be detected, and in addition, the damage description information of the reference pipe section can objectively and accurately reflect the damage condition of the pipe section to be detected under the action of geological disasters.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an early warning method for vulnerability of an oil and gas pipeline under the action of geological disasters;
FIG. 2 is a flow chart of a method for determining an impact indicator provided by the present invention;
FIG. 3 is a flow chart of a method for determining a reference pipe segment provided by the present invention;
FIG. 4 is a schematic structural diagram of an oil and gas pipeline vulnerability early warning device under the action of geological disasters;
fig. 5 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
FIG. 1 is a schematic flow chart of an early warning method for vulnerability of an oil and gas pipeline under the action of geological disasters, as shown in FIG. 1, the method comprises the following steps:
step 110, dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
step 120, determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of various influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model;
the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
Specifically, the target pipeline is first divided into a plurality of independent pipe segments to be tested. The pipeline of the same pipe section to be measured is identical in geological condition, namely geological structure, soil type and rock property, and different pipe sections to be measured can be identical or different in geological condition, namely when the pipe sections to be measured are divided, the length of the pipe sections to be measured can be flexibly controlled according to actual geographic positions.
After each independent pipe section to be detected is obtained through division, independent vulnerability prediction can be carried out on each pipe section to be detected respectively so as to predict the damage degree of each pipe section to be detected under the action of geological disasters. For any pipe section to be tested, the vulnerability prediction model corresponding to the pipe section to be tested can be determined based on the current geological disaster type (such as landslide, earthquake and the like) and the geological condition of the pipe section to be tested. It can be seen that, the combinations of different geological conditions and geological disaster types, the corresponding vulnerability prediction models are different, because the damage degree or damage mode of the pipeline by the geological disaster under the different geological conditions or geological disaster types are different, in order to more accurately predict the damage degree of each pipe section to be tested, the corresponding vulnerability prediction models can be constructed for the combinations of the various geological conditions and geological disaster types, and the prediction mode of the pipeline damage under the combinations of the geological conditions and the geological disaster types can be learned from the pipeline data of the sample pipeline under the combinations of the corresponding geological conditions and the geological disaster types, so that the accuracy of pipeline vulnerability prediction is improved.
The input corresponding to any vulnerability prediction model is an index value of each influence index of the sample pipeline/pipe section to be detected corresponding to the vulnerability prediction model, and the output at least comprises the vulnerability grade of the sample pipeline/pipe section to be detected obtained by model prediction. For any combination of geological conditions and geological disaster types, the vulnerability prediction model corresponding to the combination of the geological conditions and the geological disaster types can be trained and obtained based on index values of a plurality of sample pipelines corresponding to each influence index under the combination of the geological conditions and the geological disaster types and vulnerability grade labels of the corresponding sample pipelines. The vulnerability grade label of the sample pipeline can be determined by using a preset evaluation rule based on the actual damage condition of the corresponding sample pipeline.
Here, each influence index of any vulnerability prediction model can be obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipeline corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipeline. The disaster body indexes comprise disaster body influence directions, disaster body volumes and the like, and the pipeline indexes comprise pipeline defect density, pipeline wall thickness, internal pressure, pipes, pipe diameters, conveying media, pipeline burial depths, pipeline laying modes, pipeline relative positions and the like. When the influence indexes of the vulnerability prediction model are set, indexes which are more important for predicting the vulnerability level are screened from the disaster indexes and the pipeline indexes to serve as the influence indexes, so that the prediction capacity of the corresponding model is improved. The index with higher importance is obtained by screening after importance analysis is carried out on the disaster body index and the pipeline index based on the index value of the sample pipeline corresponding to each disaster body index and the pipeline index and the vulnerability grade label of the corresponding sample pipeline, so that the index with higher importance is used as an influence index, the index which is more effective for predicting the pipeline vulnerability can be screened more objectively and more accurately, and the influence of subjective experience restriction when the index is selected manually is avoided. It should be noted that, the geological condition and the type of the facing geological disaster of the sample pipeline according to which each influence index of the vulnerability prediction model is determined are consistent with the geological condition and the type of the geological disaster corresponding to the vulnerability prediction model.
In some embodiments, as shown in fig. 2, the respective impact indicators of the vulnerability prediction model corresponding to any pipe segment under test may be determined based on the following steps:
step 210, performing iterative training on the initial feature selection model based on index values of a plurality of sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines;
the type of geological disaster faced by the sample pipeline and the geological conditions of the sample pipeline are the same as those of any pipe section to be tested; the characteristic selection model and the vulnerability prediction model corresponding to any pipe section to be tested have the same structure except that the number of neurons of an input layer is different, and the neurons of the input layer in the characteristic selection model correspond to disaster indexes and pipeline indexes one by one; neurons of an input layer in the vulnerability prediction model correspond to the influence indexes one by one;
step 220, determining a first importance degree of the disaster index and the pipeline index based on weights corresponding to each neuron of the input layer in the feature selection model after iterative training;
step 230, screening the impact index from the disaster index and the pipeline index based on the first importance degree of the disaster index and the pipeline index.
Specifically, for any vulnerability prediction model, in order to more accurately and objectively screen out an influence index that truly affects the pipeline vulnerability, a feature selection model with a structure similar to that of the vulnerability prediction model can be constructed. The characteristic selection model and the vulnerability prediction model can be BP neural networks, the two models have the same structure except that the quantity of neurons of the input layer is different, the neurons of the input layer in the characteristic selection model correspond to the disaster indexes and the pipeline indexes one by one, and the neurons of the input layer in the vulnerability prediction model correspond to the influence indexes one by one. That is, the input layer of the feature selection model/vulnerability prediction model, when receiving the index values of the respective indices, will match the order of the input data with the order of the neurons corresponding to the respective indices. It should be noted that, the feature selection model is constructed before the vulnerability prediction model, and the vulnerability prediction model is constructed formally after the influence indexes are determined.
And then, performing iterative training on the initial feature selection model based on index values of the plurality of sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines. The type of the geological disaster faced by the sample pipeline and the geological condition of the sample pipeline are the same as the type of the geological disaster and the geological condition corresponding to the vulnerability prediction model, and the number of iterative training can be set based on the actual training situation, for example, 100epoch can be set. After the iterative training is finished, the neurons of each layer in the feature selection model learn corresponding weights, and the weights corresponding to the neurons of the input layers can represent the importance of disaster indexes and pipeline indexes corresponding to the neurons in the model prediction process. Because the feature selection model and the vulnerability prediction model under the same geological condition and geological disaster type combination have the same structure except the different quantity of neurons of the input layer, the important neurons in the feature selection model also have importance in the vulnerability prediction model. Therefore, the first importance degree of each disaster index and the pipeline index can be determined based on the weight corresponding to each neuron of the input layer in the feature selection model after the iterative training is finished, and the influence index of the vulnerability prediction model is accurately screened from each disaster index and the pipeline index based on the first importance degree of each disaster index and the pipeline index.
In other embodiments, in order to screen the impact indicators from the disaster indexes and the pipeline indexes based on the first importance degrees of the disaster indexes and the pipeline indexes, the feature selection model after the iterative training may be utilized to sequentially input the index values of each sample pipeline corresponding to each disaster index and the pipeline index into the feature selection model to predict, so as to obtain the first probability of each sample pipeline corresponding to each vulnerability level output by the model. In addition, a disaster index and/or a pipeline index with the first importance degree being lower than a first preset threshold value can be selected as candidate indexes, and a disaster index and/or a pipeline index with the first importance degree being greater than or equal to the first preset threshold value can be selected as an influence index. Although the first importance level of the candidate index is relatively low, the candidate index cannot be completely considered to be unimportant in determining the pipeline vulnerability, so in order to further improve the screening accuracy of the influence index and further ensure the accuracy of pipeline vulnerability prediction, further screening can be performed in the candidate index so as to avoid missing important indexes.
Specifically, after randomizing the index values of each sample pipeline corresponding to any candidate index, the feature selection model after the iterative training is used again to predict the index values of each sample pipeline corresponding to each disaster body index and pipeline index, so as to obtain the second probability of each sample pipeline corresponding to each vulnerability level. Since the index value of each sample pipe corresponding to the candidate index is randomized, the influence of the candidate index is significantly weakened when the feature selection model predicts the vulnerability level of each sample pipe. If the prediction result of the feature selection model is obviously changed after weakening the influence of the candidate index in the prediction process, the candidate index is shown to have certain importance for the pipeline vulnerability prediction process of the feature selection model. Thus, a second degree of importance of the candidate indicator may be determined based on a difference between the first probability and the second probability for each sample pipe corresponding to each vulnerability class, and the impact indicator may be screened from each candidate indicator based on the second degree of importance of each candidate indicator.
Here, the difference between the first probability and the second probability of each vulnerability level corresponding to any sample pipe may be summed to obtain the feature loss corresponding to the sample pipe, and the feature loss corresponding to each sample pipe may be summed to obtain the second importance degree of the candidate index. Then, based on the first importance level and the second importance level of any candidate index, the comprehensive importance level of the candidate index is determined. For example, the first importance level and the second importance level of the candidate index may be normalized, and then the normalized values may be added to obtain the overall importance level of the candidate index. If the comprehensive importance degree of the candidate index is greater than a second preset threshold value, the candidate index can be determined to be an influence index.
After the trained vulnerability prediction model corresponding to any pipe section to be detected is determined, the index values of all the influence indexes of the pipe section to be detected corresponding to the vulnerability prediction model are input into the vulnerability prediction model, and the vulnerability grade of the pipe section to be detected and the damage description information of the reference pipe section output by the vulnerability prediction model are obtained. The reference pipe section is a pipe section with a vulnerability grade, a geological disaster type and known damage condition, wherein the geological condition of the pipe section is the same as that of the pipe section to be detected, and the pipeline state of any pipe section is determined based on the index value of the pipe section corresponding to the pipeline influence index (namely, the index related to the attribute of the pipeline in the influence index) in the influence index of the vulnerability prediction model. The damage description information of the reference pipe section is pushed to the user by determining the reference pipe section corresponding to the pipe section to be detected, so that the user can more intuitively know the damage condition of the pipe section to be detected. In addition, it should be noted that the vulnerability level of the reference pipe section is the same as that of the pipe section to be measured, that is, the reference pipe section is a pipe section with the same vulnerability level as that of the pipe section to be measured, which is predicted by the same pipe vulnerability prediction mode, and the type of the geological disaster facing the reference pipe section, the geological condition of the reference pipe section are the same as those of the pipe section to be measured, and the pipe state of the reference pipe section is the closest to that of the pipe section to be measured, so that the damage description information of the reference pipe section can objectively and accurately reflect the damage condition of the pipe section to be measured under the action of the geological disaster.
In some embodiments, as shown in fig. 3, the reference pipe segment may be determined based on the following steps:
step 310, obtaining the latest index values of the corresponding influence indexes of the known pipe sections which are in the same type of the geological disaster, are in the same geological condition as any pipe section to be tested and have known damage conditions before the occurrence of the geological disaster;
step 320, inputting the latest index values of each known pipe section corresponding to each influence index into the vulnerability prediction model respectively to obtain vulnerability levels of each known pipe section output by the vulnerability prediction model;
step 330, selecting a known pipe section with the same vulnerability level as any pipe section to be tested as a candidate pipe section;
step 340, determining a pipeline state vector of the candidate pipe section based on the latest index value of the pipeline influence index in the influence indexes corresponding to the candidate pipe section, and determining the pipeline state vector of any pipe section to be tested based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section to be tested; the pipeline state vector characterizes the pipeline state of the corresponding pipeline section;
and step 350, screening the reference pipe section from the candidate pipe sections based on the distance between the pipeline state vector of any pipe section to be tested and the pipeline state vector of each candidate pipe section.
Specifically, the type of the geological disaster faced, the known pipe section which is the same as the pipe section to be measured in the geological condition and has known damage condition, and the latest index value of the known pipe section corresponding to each influence index before the occurrence of the geological disaster can be obtained from the database. And then, sequentially inputting the latest index values of the influence indexes corresponding to the known pipe sections into the vulnerability prediction model corresponding to the pipe section to be detected, and obtaining the vulnerability grade of the known pipe sections output by the vulnerability prediction model. Namely, for any known pipe section, inputting the latest index value of each influence index corresponding to the known pipe section into the vulnerability prediction model for prediction to obtain the vulnerability grade of the known pipe section output by the model. Therefore, if the vulnerability level of the known pipe section is consistent with the vulnerability level of the pipe section to be measured, it can be determined that the damage degree of the pipe section to be measured is objectively similar to that of the corresponding known pipe section.
In order to more intuitively and accurately show the possible damage condition of the pipe section to be tested to the user, the known pipe section with the same vulnerability level as the pipe section to be tested can be selected as a candidate pipe section. And then, determining the pipeline state vector of the candidate pipeline section based on the latest index value of the pipeline influence index in the corresponding influence indexes of the candidate pipeline section, and determining the pipeline state vector of the pipeline section to be tested based on the index value of the pipeline influence index in the corresponding influence indexes of the pipeline section to be tested. Wherein the pipeline state vector of any pipe section characterizes the pipeline state of the corresponding pipe section, and whether the pipeline states of the two pipe sections are similar can be determined according to the pipeline state vectors of the two pipe sections. Therefore, the distance between the pipeline state vector of the pipe section to be measured and the pipeline state vector of each candidate pipe section can be calculated, and then a plurality of candidate pipe sections with the smallest distance are selected from the candidate pipe sections as reference pipe sections.
In some embodiments, when determining the pipeline state vector of the candidate pipeline segment or the pipeline segment to be measured, the index values of the pipeline impact indexes in the impact indexes corresponding to the candidate pipeline segment or the pipeline segment to be measured may be normalized respectively to obtain normalized index values corresponding to the pipeline impact indexes of the corresponding pipeline segment. And then, determining the weight of each pipeline influence index based on the weight of each neuron of the input layer in the vulnerability prediction model, and weighting and combining the normalized index values of the candidate pipeline section or the pipeline section to be detected corresponding to each pipeline influence index based on the weight of each pipeline influence index to obtain the pipeline state vector of the candidate pipeline section or the pipeline section to be detected. Assuming that the normalized index values of the candidate pipe section or the pipe section to be tested corresponding to the pipe impact indexes are p1, p2, p3, and pn, and the weights of the pipe impact indexes are w1, w2, w3, and wn, respectively, the pipe state vector of the candidate pipe section or the pipe section to be tested may be (p1×w1, p2×w2, p3×w3, and pn×wn).
In some embodiments, after the actual damage condition of each pipe segment to be tested in the target pipeline is obtained, the vulnerability level label of each pipe segment to be tested may be set based on the actual damage condition of each pipe segment to be tested, and fine tuning is performed on the vulnerability prediction model corresponding to each pipe segment to be tested based on the vulnerability level label of each pipe segment to be tested.
According to the method provided by the embodiment of the invention, the accuracy of pipeline vulnerability prediction can be improved by constructing the corresponding vulnerability prediction model for different geological conditions and geological disaster type combinations and learning the prediction mode of pipeline damage under the geological conditions and geological disaster type combinations from the pipeline data of the sample pipeline under the corresponding geological conditions and geological disaster type combinations; meanwhile, index values corresponding to each disaster body index and pipeline index and vulnerability grade labels of corresponding sample pipelines are based on the sample pipelines, indexes with higher importance are obtained by screening after importance analysis is carried out on the disaster body index and the pipeline index, and are used as influence indexes, so that indexes which are more effective in predicting pipeline vulnerability can be screened more objectively and more accurately, and influence of subjective experience restriction when indexes are selected manually is avoided; after the trained vulnerability prediction model corresponding to any pipe section to be detected is determined, the index value of each influence index of the pipe section to be detected corresponding to the vulnerability prediction model is input into the vulnerability prediction model, so that the vulnerability grade of the pipe section to be detected and the damage description information of the reference pipe section output by the vulnerability prediction model are obtained, a user can more intuitively know the damage condition of the pipe section to be detected, and in addition, the damage description information of the reference pipe section can objectively and accurately reflect the damage condition of the pipe section to be detected under the action of geological disasters.
The oil and gas pipeline vulnerability early warning device under the action of geological disasters provided by the invention is described below, and the oil and gas pipeline vulnerability early warning device under the action of geological disasters described below and the oil and gas pipeline vulnerability early warning method under the action of geological disasters described above can be correspondingly referred to each other.
Based on any of the above embodiments, fig. 4 is a schematic structural diagram of an early warning device for vulnerability of an oil and gas pipeline under the action of geological disasters, as shown in fig. 4, the device includes:
a pipe section dividing unit 410 for dividing the target pipeline into a plurality of independent pipe sections to be measured; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
the vulnerability prediction unit 420 is configured to determine, for any pipe segment to be tested, a vulnerability prediction model corresponding to the any pipe segment to be tested based on a current geological disaster type and geological conditions where the any pipe segment to be tested is located, and input index values of respective impact indexes of the any pipe segment to be tested corresponding to the vulnerability prediction model into the vulnerability prediction model, so as to obtain a vulnerability grade of the any pipe segment to be tested and damage description information of a reference pipe segment output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
According to the device provided by the embodiment of the invention, the corresponding vulnerability prediction model is constructed for different geological conditions and geological disaster type combinations, and the prediction mode of the pipeline damage under the geological conditions and geological disaster type combinations is learned from the pipeline data of the sample pipeline under the corresponding geological conditions and geological disaster type combinations, so that the accuracy of pipeline vulnerability prediction can be improved; meanwhile, index values corresponding to each disaster body index and pipeline index and vulnerability grade labels of corresponding sample pipelines are based on the sample pipelines, indexes with higher importance are obtained by screening after importance analysis is carried out on the disaster body index and the pipeline index, and are used as influence indexes, so that indexes which are more effective in predicting pipeline vulnerability can be screened more objectively and more accurately, and influence of subjective experience restriction when indexes are selected manually is avoided; after the trained vulnerability prediction model corresponding to any pipe section to be detected is determined, the index value of each influence index of the pipe section to be detected corresponding to the vulnerability prediction model is input into the vulnerability prediction model, so that the vulnerability grade of the pipe section to be detected and the damage description information of the reference pipe section output by the vulnerability prediction model are obtained, a user can more intuitively know the damage condition of the pipe section to be detected, and in addition, the damage description information of the reference pipe section can objectively and accurately reflect the damage condition of the pipe section to be detected under the action of geological disasters.
Based on any one of the above embodiments, each impact index of the vulnerability prediction model corresponding to any one pipe section to be tested is determined based on the following steps:
iterative training is carried out on the initial feature selection model based on index values of a plurality of sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines; the type of geological disaster faced by the sample pipeline and the geological conditions of the sample pipeline are the same as those of any pipe section to be tested; the characteristic selection model and the vulnerability prediction model corresponding to any pipe section to be tested have the same structure except that the number of neurons of an input layer is different, and the neurons of the input layer in the characteristic selection model correspond to disaster indexes and pipeline indexes one by one; neurons of an input layer in the vulnerability prediction model correspond to the influence indexes one by one;
determining a first importance degree of the disaster index and the pipeline index based on weights corresponding to each neuron of the input layer in the feature selection model after iterative training;
and screening influence indexes from the disaster body indexes and the pipeline indexes based on the first importance degrees of the disaster body indexes and the pipeline indexes.
Based on any one of the foregoing embodiments, the screening, based on the first importance levels of the disaster body index and the pipeline index, the impact index from the disaster body index and the pipeline index specifically includes:
predicting the index values of the disaster indexes and the pipeline indexes corresponding to the sample pipelines by utilizing the feature selection model after the iterative training, so as to obtain a first probability of the vulnerability grades corresponding to the sample pipelines;
the disaster body index and/or the pipeline index with the first importance degree being lower than a first preset threshold value are selected as candidate indexes, and the disaster body index and/or the pipeline index with the first importance degree being greater than or equal to the first preset threshold value are selected as influence indexes;
carrying out randomization treatment on index values of each sample pipeline corresponding to any candidate index, and then, utilizing the feature selection model after iterative training, predicting the index values of each sample pipeline corresponding to each disaster body index and pipeline index to obtain a second probability of each sample pipeline corresponding to each vulnerability grade;
determining a second importance degree of any candidate index based on a difference between the first probability and the second probability of each sample pipe corresponding to each vulnerability class;
And screening the candidate indexes for influencing indexes based on the second importance degree of the candidate indexes.
Based on any one of the foregoing embodiments, the determining the second importance of any one candidate indicator based on the difference between the first probability and the second probability of each sample pipe corresponding to each vulnerability class specifically includes:
summing the differences between the first probability and the second probability of each vulnerability grade corresponding to any sample pipeline to obtain the corresponding characteristic loss of any sample pipeline;
and summing the characteristic losses corresponding to the sample pipelines to obtain a second importance degree of any candidate index.
Based on any one of the foregoing embodiments, the screening the impact indicator from the candidate indicators based on the second importance degree of the candidate indicators specifically includes:
determining the comprehensive importance degree of any candidate index based on the first importance degree and the second importance degree of the any candidate index aiming at the any candidate index;
and if the comprehensive importance degree of any candidate index is greater than a second preset threshold value, determining the any candidate index as an influence index.
Based on any of the above embodiments, the reference pipe segment is determined based on the steps of:
Acquiring the latest index values of the corresponding influence indexes of the known pipe sections which are of the same type, are in the same geological condition as any pipe section to be tested and have known damage conditions before the occurrence of the geological disaster;
inputting the latest index values of the known pipe sections corresponding to the influence indexes into the vulnerability prediction model respectively to obtain vulnerability levels of the known pipe sections output by the vulnerability prediction model;
selecting a known pipe section with the vulnerability grade identical to that of any pipe section to be tested as a candidate pipe section;
determining a pipeline state vector of the candidate pipe section based on the latest index value of the pipeline influence index in the influence indexes corresponding to the candidate pipe section, and determining the pipeline state vector of any pipe section to be tested based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section to be tested; the pipeline state vector characterizes the pipeline state of the corresponding pipeline section;
and screening the reference pipe section from the candidate pipe sections based on the distance between the pipeline state vector of any pipe section to be tested and the pipeline state vector of each candidate pipe section.
Based on any of the above embodiments, the pipe state vector for the candidate pipe segment or any of the pipe segments under test is determined based on the steps of:
Respectively normalizing index values of the pipeline influence indexes in the candidate pipeline sections or any pipeline section to be tested corresponding to the influence indexes to obtain normalized index values of the pipeline influence indexes corresponding to the corresponding pipeline sections;
determining the weight of each pipeline influence index based on the weight of each neuron of the input layer in the vulnerability prediction model;
and weighting and combining the normalized index values of the candidate pipe section or any pipe section to be tested corresponding to the pipeline influence indexes based on the weight of the pipeline influence indexes to obtain the pipeline state vector of the candidate pipe section or any pipe section to be tested.
Based on any of the above embodiments, the vulnerability prediction model is trained based on index values of the plurality of sample pipes corresponding to respective impact indexes and vulnerability grade labels of the corresponding sample pipes.
Based on any of the above embodiments, the method further includes a trimming unit for:
and after the actual damage condition of each pipe section to be detected in the target pipeline is obtained, fine-tuning the vulnerability prediction model based on the actual damage condition of each pipe section to be detected.
Fig. 5 is a schematic structural diagram of an electronic device according to the present invention, and as shown in fig. 5, the electronic device may include: processor 510, memory 520, communication interface (Communications Interface) 530, and communication bus 540, wherein processor 510, memory 520, and communication interface 530 communicate with each other via communication bus 540. Processor 510 may invoke logic instructions in memory 520 to perform a method for pre-warning of vulnerability of an oil and gas pipeline under the action of a geological disaster, the method comprising: dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties; determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section; the vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
Further, the logic instructions in the memory 520 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of pre-warning of vulnerability of an oil and gas pipeline under the effect of geological disasters provided by the above methods, the method comprising: dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties; determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section; the vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
In still another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, is implemented to perform the above-provided method for early warning of vulnerability of an oil and gas pipeline under the action of geological disasters, the method comprising: dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties; determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section; the vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The utility model provides an oil gas pipeline vulnerability early warning method under geological disaster effect which is characterized in that the method comprises the following steps:
dividing a target pipeline into a plurality of independent pipe sections to be tested; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
determining a vulnerability prediction model corresponding to any pipe section to be tested according to the current geological disaster type and geological conditions of the any pipe section to be tested, and inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be tested into the vulnerability prediction model to obtain vulnerability grade of the any pipe section to be tested and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
2. The method for early warning of vulnerability of oil and gas pipelines under geological disaster action according to claim 1, wherein each influence index of the vulnerability prediction model corresponding to any pipe section to be tested is determined based on the following steps:
iterative training is carried out on the initial feature selection model based on index values of a plurality of sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines; the type of geological disaster faced by the sample pipeline and the geological conditions of the sample pipeline are the same as those of any pipe section to be tested; the characteristic selection model and the vulnerability prediction model corresponding to any pipe section to be tested have the same structure except that the number of neurons of an input layer is different, and the neurons of the input layer in the characteristic selection model correspond to disaster indexes and pipeline indexes one by one; neurons of an input layer in the vulnerability prediction model correspond to the influence indexes one by one;
Determining a first importance degree of the disaster index and the pipeline index based on weights corresponding to each neuron of the input layer in the feature selection model after iterative training;
and screening influence indexes from the disaster body indexes and the pipeline indexes based on the first importance degrees of the disaster body indexes and the pipeline indexes.
3. The method for early warning vulnerability of oil and gas pipelines under geological disaster action according to claim 2, wherein the screening of the impact indicators from the disaster indicators and the pipeline indicators based on the first importance levels of the disaster indicators and the pipeline indicators specifically comprises:
predicting the index values of the disaster indexes and the pipeline indexes corresponding to the sample pipelines by utilizing the feature selection model after the iterative training, so as to obtain a first probability of the vulnerability grades corresponding to the sample pipelines;
the disaster body index and/or the pipeline index with the first importance degree being lower than a first preset threshold value are selected as candidate indexes, and the disaster body index and/or the pipeline index with the first importance degree being greater than or equal to the first preset threshold value are selected as influence indexes;
carrying out randomization treatment on index values of each sample pipeline corresponding to any candidate index, and then, utilizing the feature selection model after iterative training, predicting the index values of each sample pipeline corresponding to each disaster body index and pipeline index to obtain a second probability of each sample pipeline corresponding to each vulnerability grade;
Determining a second importance degree of any candidate index based on a difference between the first probability and the second probability of each sample pipe corresponding to each vulnerability class;
and screening the candidate indexes for influencing indexes based on the second importance degree of the candidate indexes.
4. The method for early warning vulnerability of oil and gas pipelines under geological disaster action according to claim 3, wherein the determining the second importance degree of any candidate index based on the difference between the first probability and the second probability of each sample pipeline corresponding to each vulnerability level specifically comprises:
summing the differences between the first probability and the second probability of each vulnerability grade corresponding to any sample pipeline to obtain the corresponding characteristic loss of any sample pipeline;
and summing the characteristic losses corresponding to the sample pipelines to obtain a second importance degree of any candidate index.
5. The method for early warning of vulnerability of oil and gas pipelines under geological disaster action according to claim 3, wherein the screening the candidate indexes for influencing indexes based on the second importance degree of the candidate indexes specifically comprises the following steps:
determining the comprehensive importance degree of any candidate index based on the first importance degree and the second importance degree of the any candidate index aiming at the any candidate index;
And if the comprehensive importance degree of any candidate index is greater than a second preset threshold value, determining the any candidate index as an influence index.
6. The method for early warning of vulnerability of oil and gas pipelines under geological disasters according to claim 2, wherein the reference pipe section is determined based on the following steps:
acquiring the latest index values of the corresponding influence indexes of the known pipe sections which are of the same type, are in the same geological condition as any pipe section to be tested and have known damage conditions before the occurrence of the geological disaster;
inputting the latest index values of the known pipe sections corresponding to the influence indexes into the vulnerability prediction model respectively to obtain vulnerability levels of the known pipe sections output by the vulnerability prediction model;
selecting a known pipe section with the vulnerability grade identical to that of any pipe section to be tested as a candidate pipe section;
determining a pipeline state vector of the candidate pipe section based on the latest index value of the pipeline influence index in the influence indexes corresponding to the candidate pipe section, and determining the pipeline state vector of any pipe section to be tested based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section to be tested; the pipeline state vector characterizes the pipeline state of the corresponding pipeline section;
And screening the reference pipe section from the candidate pipe sections based on the distance between the pipeline state vector of any pipe section to be tested and the pipeline state vector of each candidate pipe section.
7. The method of claim 6, wherein the pipeline status vector for the candidate pipe segment or any of the candidate pipe segments is determined based on the steps of:
respectively normalizing index values of the pipeline influence indexes in the candidate pipeline sections or any pipeline section to be tested corresponding to the influence indexes to obtain normalized index values of the pipeline influence indexes corresponding to the corresponding pipeline sections;
determining the weight of each pipeline influence index based on the weight of each neuron of the input layer in the vulnerability prediction model;
and weighting and combining the normalized index values of the candidate pipe section or any pipe section to be tested corresponding to the pipeline influence indexes based on the weight of the pipeline influence indexes to obtain the pipeline state vector of the candidate pipe section or any pipe section to be tested.
8. The method for early warning of vulnerability of oil and gas pipelines under geological disaster action according to claim 1, wherein the vulnerability prediction model is trained based on index values of a plurality of sample pipelines corresponding to each influence index and vulnerability grade labels of the corresponding sample pipelines.
9. The method for early warning of vulnerability of oil and gas pipelines under the action of geological disasters according to claim 1, further comprising:
and after the actual damage condition of each pipe section to be detected in the target pipeline is obtained, fine-tuning the vulnerability prediction model based on the actual damage condition of each pipe section to be detected.
10. An oil gas pipeline vulnerability early warning device under geological disasters effect, which is characterized by comprising:
the pipe section dividing unit is used for dividing the target pipeline into a plurality of independent pipe sections to be measured; the pipeline of the same pipe section to be detected is in the same geological condition, wherein the geological condition comprises a geological structure, a soil type and rock properties;
the vulnerability prediction unit is used for determining a vulnerability prediction model corresponding to any pipe section to be detected based on the current geological disaster type and geological conditions of the any pipe section to be detected, inputting index values of all influence indexes of the vulnerability prediction model corresponding to the any pipe section to be detected into the vulnerability prediction model, and obtaining vulnerability grade of the any pipe section to be detected and damage description information of a reference pipe section, which are output by the vulnerability prediction model; the reference pipe section is a pipe section with known damage condition, wherein the vulnerability grade, the geological disaster type and the geological conditions are the same as those of any pipe section to be tested, and the pipeline state is the closest to that of any pipe section to be tested; the pipeline state of any pipe section is determined based on the index value of the pipeline influence index in the influence indexes corresponding to any pipe section;
The vulnerability prediction model is characterized in that each influence index of the vulnerability prediction model is obtained by screening after importance analysis of disaster body indexes and pipeline indexes based on index values of the sample pipelines corresponding to each disaster body index and pipeline index and vulnerability grade labels of the corresponding sample pipelines.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311007712.1A CN116739188B (en) | 2023-08-11 | 2023-08-11 | Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311007712.1A CN116739188B (en) | 2023-08-11 | 2023-08-11 | Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116739188A CN116739188A (en) | 2023-09-12 |
CN116739188B true CN116739188B (en) | 2023-10-17 |
Family
ID=87906365
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311007712.1A Active CN116739188B (en) | 2023-08-11 | 2023-08-11 | Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116739188B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109409719A (en) * | 2018-10-17 | 2019-03-01 | 西南石油大学 | A kind of lower long-distance oil & gas pipeline vulnerability assessment method of landslide effect |
CN115081287A (en) * | 2022-07-01 | 2022-09-20 | 西南石油大学 | Method for evaluating gas transmission pipeline under landslide action based on structural vulnerability theory |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102021114584A1 (en) * | 2021-06-07 | 2022-12-08 | Endress+Hauser Group Services Ag | Method of monitoring a condition within a piping system for degradation by deposits, abrasion or corrosion |
-
2023
- 2023-08-11 CN CN202311007712.1A patent/CN116739188B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109409719A (en) * | 2018-10-17 | 2019-03-01 | 西南石油大学 | A kind of lower long-distance oil & gas pipeline vulnerability assessment method of landslide effect |
CN115081287A (en) * | 2022-07-01 | 2022-09-20 | 西南石油大学 | Method for evaluating gas transmission pipeline under landslide action based on structural vulnerability theory |
Also Published As
Publication number | Publication date |
---|---|
CN116739188A (en) | 2023-09-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3671201B1 (en) | An improved method for evaluating pipe condition | |
CN109922032B (en) | Method, device, equipment and storage medium for determining risk of logging in account | |
CN112819207B (en) | Geological disaster space prediction method, system and storage medium based on similarity measurement | |
CN111737642A (en) | Comprehensive evaluation method for failure risk of submarine pipeline based on fuzzy network analysis method | |
Neves et al. | An approach to decision‐making analysis for implementation of structural health monitoring in bridges | |
Jeong et al. | Prediction of flexible pavement deterioration in relation to climate change using fuzzy logic | |
KR102061500B1 (en) | System predicting tsunami based on artificial neural network | |
CN115270527B (en) | Real-time assessment method, equipment and storage medium for road collapse risk | |
Li et al. | Real‐Time Warning and Risk Assessment of Tailings Dam Disaster Status Based on Dynamic Hierarchy‐Grey Relation Analysis | |
Parvizsedghy et al. | Consequence of failure: Neurofuzzy-based prediction model for gas pipelines | |
CN116415481A (en) | Regional landslide hazard risk prediction method and device, computer equipment and storage medium | |
JP2016029560A (en) | Decision support system, decision support method, and non-temporary storage medium for assessing intervention action | |
CN111666684B (en) | Circumferential weld risk prediction method and device for conveying pipeline and readable storage medium | |
CN116739188B (en) | Early warning method and device for vulnerability of oil and gas pipeline under geological disaster effect | |
CN115729761B (en) | Hard disk fault prediction method, system, equipment and medium | |
CN116502804A (en) | Maintenance guarantee point site selection method | |
CN115906663A (en) | Building safety evaluation model establishing method, evaluation method, server and system | |
CN115829209A (en) | Environment-friendly intelligent warehouse environment-friendly quality analysis method and device based on carbon path | |
CN115358454A (en) | Coal and gas outburst prediction method based on extension-fuzzy hierarchical analysis theory | |
St. Clair et al. | Development of a fuzzy inference performance index for ferrous drinking water pipelines | |
CN111667133B (en) | Comprehensive evaluation method and system for integrity of pipeline | |
Atique | Analysis of urban pipe deterioration using Copula method | |
Morato Dominguez | Optimal Inspection and Maintenance Planning for Deteriorating Structures via Markov Decision Processes and Deep Reinforcement Learning. Application to Offshore Wind Substructures. | |
CN115931264A (en) | Gas pipeline evaluation method and device | |
Cheng et al. | Data-Driven Machine Learning Techniques Case Study for Corrosion Depth Prediction in Facilities Station Piping |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |