Oil and gas pipeline risk assessment method and system combining static and dynamic analysis

Abstract

The application provides an oil and gas pipeline risk assessment method and system fusing static and dynamic analysis, and belongs to the field of oil and gas pipeline risk assessment. The method comprises the following steps: obtaining a plurality of static evaluation parameters and a plurality of dynamic evaluation parameter sets; performing mutual exclusion analysis to obtain a static mutual exclusion coefficient and a plurality of dynamic mutual exclusion coefficients; respectively performing risk analysis to obtain a static risk parameter and a plurality of dynamic risk parameters; and performing pipeline control scheme optimization according to the static risk parameter and the plurality of dynamic risk parameters, and the static mutual exclusion coefficient and the plurality of dynamic mutual exclusion coefficients to obtain an optimal pipeline control scheme and perform risk control. The application solves the technical problem that the existing oil and gas pipeline risk assessment only relies on static evaluation, resulting in inaccurate risk assessment and weak pertinence of the pipeline control scheme, and achieves the technical effect of eliminating evaluation index conflicts through static evaluation and dynamic analysis and mutual exclusion analysis to optimize the pipeline control scheme, thereby improving the accuracy of risk assessment and the effect of control.

Description

Technical Field

[0001] This invention relates to the field of oil and gas pipeline risk assessment, and in particular to a method and system for oil and gas pipeline risk assessment that integrates static and dynamic analysis. Background Technology

[0002] As vital energy transportation infrastructure, the safe operation of oil and gas pipelines is directly related to energy security and public safety. With the continuous expansion of oil and gas pipeline construction and the increase in their service life, risk assessment and safety management of oil and gas pipelines have become key issues in pipeline operation and management.

[0003] Current pipeline risk assessment technologies rely on static evaluation, which makes risk judgments based on relatively fixed factors such as pipeline design parameters, construction quality, and operating environment. This results in inaccurate risk assessment results that cannot accurately reflect the actual risk status of the pipeline, thereby affecting the pertinence and effectiveness of the control plan. Summary of the Invention

[0004] This invention addresses the technical problem that existing oil and gas pipeline risk assessments rely solely on static evaluations, resulting in inaccurate risk assessments and weakly targeted control measures. It provides an oil and gas pipeline risk assessment method and system that integrates static and dynamic analysis to solve this problem.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides a risk assessment method for oil and gas pipelines that integrates static and dynamic analysis, comprising: acquiring multiple evaluation indicators for risk assessment of a target oil and gas pipeline, acquiring multiple static evaluation parameters for the multiple evaluation indicators, and collecting multiple sets of dynamic evaluation parameters; performing mutual exclusion analysis on the multiple static evaluation parameters and the multiple sets of dynamic evaluation parameters to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients; performing risk analysis on the multiple static evaluation parameters and the multiple sets of dynamic evaluation parameters respectively to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results; and optimizing the control scheme using the VIKOR algorithm based on the static risk parameters, the multiple dynamic risk parameters, the static mutual exclusion coefficients, and the multiple dynamic mutual exclusion coefficients to obtain the optimal control scheme for risk control.

[0007] Secondly, this invention provides an oil and gas pipeline risk assessment system integrating static and dynamic analysis, comprising: a data acquisition module, used to acquire multiple evaluation indicators for risk assessment of the target oil and gas pipeline, acquire multiple static evaluation parameters of the multiple evaluation indicators, and collect multiple sets of dynamic evaluation parameters; a mutual exclusion analysis module, used to perform mutual exclusion analysis on the multiple static evaluation parameters and the multiple sets of dynamic evaluation parameters to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients; a risk analysis module, used to perform risk analysis based on the multiple static evaluation parameters and the multiple sets of dynamic evaluation parameters respectively to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results; and a scheme optimization module, used to optimize the control scheme using the VIKOR algorithm based on the static risk parameters, the multiple dynamic risk parameters, the static mutual exclusion coefficients, and the multiple dynamic mutual exclusion coefficients to obtain the optimal control scheme for risk control.

[0008] The beneficial effects of this invention are:

[0009] This study acquires multiple evaluation indicators for risk assessment of target oil and gas pipelines, along with multiple static evaluation parameters and dynamic evaluation parameter sets. Simultaneous acquisition of both static and dynamic evaluation parameter sets provides a data foundation for subsequent fusion analysis, ensuring that risk assessment comprehensively considers both static evaluation and real-time status. Mutual exclusion analysis is performed on the multiple static and dynamic evaluation parameter sets to obtain static and dynamic mutual exclusion coefficients. By calculating and quantifying the differences between evaluation parameters into mutual exclusion coefficients, conflicting relationships between different evaluation indicators are effectively identified and quantified, laying the foundation for eliminating evaluation bias. Based on the multiple static and dynamic evaluation parameters... The evaluation parameter set is used to conduct risk analysis, obtaining static risk parameters and multiple dynamic risk parameters as risk assessment results. By analyzing the risks of oil and gas pipelines under static and dynamic evaluation parameters respectively, multi-dimensional risk assessment results are obtained, improving the comprehensiveness and accuracy of risk identification. Based on the static risk parameters, multiple dynamic risk parameters, static mutual exclusion coefficients, and multiple dynamic mutual exclusion coefficients, the VIKOR algorithm is used to optimize the control scheme, obtaining the optimal control scheme for risk management. By comprehensively considering risk parameters and mutual exclusion coefficients through the VIKOR algorithm, the overall effectiveness is improved while reducing individual regrets, thus optimizing the control scheme and improving the pertinence and effectiveness of risk management.

[0010] The above technical solution integrates static assessment and dynamic analysis, overcoming the limitations of static evaluation methods. Through mutual exclusion analysis, it effectively identifies and eliminates conflicting relationships between different evaluation indicators, avoiding evaluation bias. Furthermore, it uses the VIKOR algorithm to comprehensively optimize the control scheme, improving overall control effectiveness while reducing individual shortcomings. This significantly enhances the accuracy of oil and gas pipeline risk assessment and the relevance of control schemes, thereby improving the overall control effect. Attached Figure Description

[0011] Figure 1 A flowchart illustrating the oil and gas pipeline risk assessment method that integrates static and dynamic analysis provided by this invention;

[0012] Figure 2 A schematic diagram of the structure of the oil and gas pipeline risk assessment system that integrates static and dynamic analysis provided by the present invention.

[0013] In the attached diagram, the components represented by each number are as follows:

[0014] Data acquisition module 11, mutual exclusion analysis module 12, risk analysis module 13, and solution optimization module 14. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0017] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0018] Example 1, as Figure 1 As shown, embodiments of the present invention provide a method for risk assessment of oil and gas pipelines that integrates static and dynamic analysis, including:

[0019] S1. Obtain multiple evaluation indicators for risk assessment of the target oil and gas pipeline, obtain multiple static evaluation parameters for multiple evaluation indicators, and collect multiple sets of dynamic evaluation parameters.

[0020] Specifically, a target oil and gas pipeline refers to a specific oil and gas transportation pipeline that requires risk assessment and management. This target oil and gas pipeline can be a crude oil pipeline, a natural gas pipeline, or a refined oil pipeline. First, multiple evaluation indicators are obtained for the risk assessment of the target oil and gas pipeline. These indicators include pipeline-related indicators, environmental factor indicators, and operational status indicators. Pipeline-related indicators cover physical characteristics such as pipeline material, wall thickness, welding quality, and anti-corrosion coating condition. Environmental factor indicators include external impact indicators such as soil corrosivity, geological stability, and third-party construction activities. Operational status indicators cover real-time operational indicators such as operating pressure, flow rate changes, and temperature distribution.

[0021] Subsequently, multiple static evaluation parameters were collected for the target oil and gas pipeline according to various evaluation indicators. These static evaluation parameters were obtained through expert assessment. Specifically, experts evaluated technical parameters related to the pipeline itself, including material grade, wall thickness design parameters, weld joint quality grade, and anti-corrosion layer integrity. For environmental factors, they assessed environmental risk factors such as soil corrosion level, geological structure stability, frequency of third-party construction activities, and population density along the pipeline route. For operational status indicators, they assessed operational characteristic parameters such as the safety margin between pipeline design pressure and actual operating pressure, historical flow variation trends, and the rationality of temperature distribution. Based on pipeline design data, historical operating data, and inspection reports, experts used the Delphi method or analytic hierarchy process to conduct qualitative and quantitative assessments of each evaluation indicator, resulting in corresponding static evaluation parameter values ​​and multiple static evaluation parameters. These static evaluation parameters remained relatively stable over a relatively long period, reflecting the inherent risk characteristics of the target oil and gas pipeline.

[0022] Next, the target oil and gas pipeline is divided into multiple pipeline segments. Dynamic evaluation parameters for various evaluation indicators are collected using sensors to obtain multiple sets of dynamic evaluation parameters. Each pipeline segment is equipped with corresponding monitoring sensors. For pipeline body indicators, parameters such as anti-corrosion layer potential and pipe wall thickness changes are collected using corrosion monitoring probes and ultrasonic thickness gauges. For environmental factor indicators, parameters such as soil corrosion current and ground settlement displacement are collected using soil corrosion sensors and geological monitoring equipment. For operational status indicators, parameters such as operating pressure, flow rate changes, and temperature distribution are collected using pressure sensors, flow meters, and temperature sensors. Real-time data for each evaluation indicator are continuously collected at a preset sampling frequency, forming multiple sets of dynamic evaluation parameters corresponding to the static evaluation parameters.

[0023] Through the above data acquisition steps, a dual data acquisition mechanism covering static assessment and real-time monitoring was established, providing comprehensive basic data support for subsequent mutually exclusive analysis of static and dynamic data, risk assessment, and optimization of control schemes. Static evaluation parameters reflect the inherent characteristics of the pipeline, while multiple sets of dynamic evaluation parameters reflect the real-time status changes of different pipeline sections. Combining the two enables more accurate risk assessment of oil and gas pipelines.

[0024] S2. Perform mutual exclusion analysis on the multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients.

[0025] Specifically, firstly, the mean difference between multiple static evaluation parameters and multiple dynamic evaluation parameter sets is calculated to obtain the static mutual exclusion coefficient. Specifically, for each evaluation indicator, the numerical difference between the static evaluation parameter and the corresponding dynamic evaluation parameter for each pipeline segment is calculated separately; then, a weighted average is calculated for multiple difference magnitudes of the same evaluation indicator, where the weights are allocated according to the importance of each evaluation indicator; finally, the weighted difference magnitudes of all evaluation indicators are comprehensively averaged to obtain the static mutual exclusion coefficient. This static mutual exclusion coefficient reflects the overall degree of deviation between expert evaluation data and real-time monitoring data.

[0026] Next, the average difference between each dynamic evaluation parameter set and other dynamic evaluation parameter sets, as well as multiple static evaluation parameters, is calculated to obtain multiple dynamic mutual exclusion coefficients. Specifically, for each pipeline segment's dynamic evaluation parameter set, its difference from the dynamic evaluation parameter sets of other pipeline segments, and its difference from the static evaluation parameters, are calculated separately. Then, these difference values ​​are statistically averaged to obtain the dynamic mutual exclusion coefficient corresponding to that pipeline segment. Since there are multiple pipeline segments, multiple dynamic mutual exclusion coefficients are obtained, each reflecting the degree of deviation between the monitoring data of the corresponding pipeline segment and other data sources.

[0027] Through the above mutual exclusion analysis, static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients were obtained. These coefficients quantify the degree of difference between different evaluation parameters, providing a basis for data quality assessment for subsequent risk analysis and management scheme optimization. This enables the reasonable allocation of weights based on data credibility, thereby improving the accuracy of risk assessment.

[0028] S3. Based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets, conduct risk analysis to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results.

[0029] Specifically, the first step is to obtain a risk analyzer. The risk analyzer is a risk assessment model built using machine learning techniques, comprising a static risk analysis branch and a dynamic risk analysis branch. The static risk analysis branch specifically handles static evaluation parameters obtained from expert assessments, while the dynamic risk analysis branch specifically handles dynamic evaluation parameters collected by sensors. The construction process of the risk analyzer includes: collecting a set of sample static evaluation parameters, multiple sets of sample dynamic evaluation parameters, and corresponding sets of sample static and dynamic risk parameters based on historical oil and gas pipeline risk management data; then, supervised training is performed on both the static and dynamic risk analysis branches until the model converges, resulting in the risk analyzer.

[0030] Next, multiple sets of static evaluation parameters and multiple sets of dynamic evaluation parameters are input into the risk analyzer. The static risk analysis branch receives multiple static evaluation parameters as input and outputs static risk parameters, which reflect the inherent risk level of the pipeline based on expert assessment data. The dynamic risk analysis branch receives multiple sets of dynamic evaluation parameters as input, processes the dynamic evaluation parameters for each pipeline segment, and outputs multiple dynamic risk parameters, which reflect the current risk status of each pipeline segment based on real-time monitoring data.

[0031] Through the above risk analysis, static risk parameters and multiple dynamic risk parameters were obtained as the risk assessment results. The static risk parameters reflect the basic risk level of the pipeline system, while the multiple dynamic risk parameters reflect the real-time risk changes of different pipeline sections. The combination of the two provides a comprehensive risk assessment data foundation for the subsequent optimization of the VIKOR algorithm control scheme.

[0032] S4. Based on static risk parameters and multiple dynamic risk parameters, as well as static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, the VIKOR algorithm is used to optimize the control scheme, obtain the optimal control scheme, and carry out risk control.

[0033] Specifically, firstly, the risk management scheme space for oil and gas pipelines is obtained, and a first management scheme is randomly generated. This management scheme space includes various possible risk management strategies, such as maintenance plans, adjustments to inspection frequency, pressure control measures, and emergency response plans. The first management scheme serves as an initial candidate scheme, providing a starting point for subsequent iterative optimizations.

[0034] Next, the first control scheme is analyzed in conjunction with static risk parameters and multiple dynamic risk parameters, resulting in first static control scores and multiple first dynamic control scores. This process is implemented using a control predictor, which is a prediction model trained based on a set of sample control schemes, a set of sample risk parameter groups, and a set of sample control scores. The first control scheme is combined with static risk parameters and multiple dynamic risk parameters as inputs, respectively, to output first static control scores and multiple first dynamic control scores. These scores reflect the expected performance of the first control scheme under the given static risk parameters and multiple dynamic risk parameters.

[0035] Then, based on the static mutual exclusion coefficient, multiple dynamic mutual exclusion coefficients, the first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value. Specifically, the static mutual exclusion coefficient is used as the static regret weight to calculate the corresponding static utility weight; multiple dynamic mutual exclusion coefficients are used as multiple dynamic regret weights to calculate the corresponding multiple dynamic utility weights; based on the difference between each control score and the maximum control score and the corresponding weights, multiple candidate individual regret values ​​and candidate group utility values ​​are calculated; the largest candidate group utility value and the corresponding candidate individual regret value are selected as the first group utility value and the first individual regret value; finally, the first group utility value and the first individual regret value are weighted and calculated according to the fusion regret weight and the fusion utility weight to obtain the first control value.

[0036] Subsequently, the control scheme was iteratively optimized by continuously generating new candidate control schemes and calculating their corresponding control values, ultimately obtaining the optimal control scheme with the maximum control value. This optimal control scheme comprehensively considers the results of static and dynamic risk assessments as well as data reliability, achieving the dual objectives of improving overall effectiveness and reducing individual regrets, thus providing an effective risk control strategy for the target oil and gas pipeline.

[0037] Through the optimization process described above, the risk management needs of different pipeline segments can be effectively balanced in complex multi-criteria decision-making environments, avoiding the problem of neglecting overall benefits due to excessive focus on local risks in traditional methods. Simultaneously, by introducing a mutual exclusion coefficient as a weighting adjustment mechanism, the impact of risk assessment data quality on the selection of management schemes is ensured, thereby improving the accuracy and practicality of oil and gas pipeline risk management.

[0038] Furthermore, multiple evaluation indicators for risk assessment of the target oil and gas pipeline are obtained, along with multiple static evaluation parameters for these indicators, and multiple sets of dynamic evaluation parameters are collected, including:

[0039] S11. Obtain multiple evaluation indicators for risk assessment of oil and gas pipelines;

[0040] S12. Collect multiple static evaluation parameters of the target oil and gas pipeline according to multiple evaluation indicators;

[0041] S13. Divide the target oil and gas pipeline into multiple pipeline segments, and collect dynamic evaluation parameters of multiple evaluation indicators through sensors to obtain multiple sets of dynamic evaluation parameters.

[0042] In a preferred embodiment, firstly, multiple evaluation indicators for oil and gas pipeline risk assessment are obtained. These multiple evaluation indicators are the basic elements constituting the risk assessment system, including three main categories: pipeline body indicators, environmental factor indicators, and operational status indicators. Pipeline body indicators cover parameters reflecting the physical characteristics of the pipeline, such as pipeline material, wall thickness, welding quality, anti-corrosion coating condition, pipe diameter, and service life. Environmental factor indicators include external environmental risk factors such as soil corrosivity, geological stability, climate conditions, frequency of third-party construction activities, and population density. Operational status indicators cover real-time operating parameters such as operating pressure, flow rate changes, temperature distribution, vibration frequency, and pressure fluctuation amplitude. The determination of these evaluation indicators is based on expert experience to ensure the comprehensiveness of the risk assessment.

[0043] Then, multiple static evaluation parameters of the target oil and gas pipeline are collected according to various evaluation indicators. This process is achieved through expert evaluation, where experts conduct qualitative and quantitative assessments of each evaluation indicator based on technical documents such as pipeline design data, historical operating data, and periodic inspection reports. Specifically, experts use professional evaluation methods such as the Delphi method or the analytic hierarchy process to conduct technical evaluations of pipeline body indicators such as material grade and wall thickness design parameters, risk assessments of environmental factor indicators such as soil corrosion level and geological structure stability, and performance evaluations of operational status indicators such as design pressure and operating pressure safety margins. Ultimately, the corresponding static evaluation parameter values ​​are generated, resulting in multiple static evaluation parameters for the target oil and gas pipeline.

[0044] Subsequently, the target oil and gas pipeline was divided into multiple pipeline segments. Dynamic evaluation parameters for various evaluation indicators were collected using sensors, resulting in multiple sets of dynamic evaluation parameters. The segment division considered factors such as geographical location, operating conditions, and environmental conditions to ensure that each segment had relatively independent risk characteristics. Each pipeline segment was equipped with corresponding monitoring sensor equipment, including pressure sensors, temperature sensors, flow meters, vibration sensors, corrosion monitoring probes, and ultrasonic thickness gauges. These sensors continuously collected real-time data for each evaluation indicator at a preset sampling frequency, forming a time-series dynamic evaluation parameter set. Due to differences in the operating environment and conditions of different pipeline segments, the dynamic evaluation parameters of each segment exhibited different variation characteristics and numerical ranges, thus forming multiple independent sets of dynamic evaluation parameters. This provides multi-dimensional real-time monitoring data support for subsequent mutual exclusion analysis and risk assessment.

[0045] Furthermore, a mutual exclusion analysis is performed on the multiple sets of static evaluation parameters and multiple sets of dynamic evaluation parameters to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, including:

[0046] S21. Calculate the mean difference between the multiple static evaluation parameters and the multiple dynamic evaluation parameter sets to obtain the static mutual exclusion coefficient;

[0047] S22. Calculate the mean difference between each dynamic evaluation parameter set and other dynamic evaluation parameter sets as well as multiple static evaluation parameters to obtain multiple dynamic mutual exclusion coefficients.

[0048] In a preferred embodiment, firstly, the average difference between multiple sets of static evaluation parameters and multiple sets of dynamic evaluation parameters is calculated to obtain a static mutual exclusion coefficient. For each evaluation index, the deviation amplitude between the static evaluation parameters and the corresponding dynamic evaluation parameters for each pipeline segment is calculated. Specifically, for each evaluation index among pipeline body indicators, environmental factor indicators, and operational status indicators, the numerical difference between the static evaluation parameters obtained from expert evaluation and the dynamic evaluation parameters for each pipeline segment collected by sensors is calculated to obtain multiple deviation amplitudes. Then, the multiple deviation amplitudes for the same evaluation index are statistically averaged to obtain the average difference of that evaluation index. For example, assuming a target oil and gas pipeline is divided into three sections, the static evaluation parameter for the operating pressure is 8.0 MPa as assessed by experts, while the dynamic evaluation parameters for the three sections collected by sensors are 8.2 MPa, 7.9 MPa, and 8.1 MPa, respectively. The calculated deviations are |8.0 - 8.2| = 0.2, |8.0 - 7.9| = 0.1, and |8.0 - 8.1| = 0.1, respectively. The mean difference for this evaluation indicator is (0.2 + 0.1 + 0.1) / 3 = 0.133. Similarly, the mean difference for other evaluation indicators such as pipeline wall thickness, corrosion protection layer condition, and soil corrosivity is calculated. Then, the mean differences for all evaluation indicators are averaged to obtain the static mutual exclusion coefficient. This static mutual exclusion coefficient quantifies the overall degree of deviation between expert evaluation data and real-time monitoring data, reflecting the level of consistency between the two data sources.

[0049] Then, the average difference between each dynamic evaluation parameter set and other dynamic evaluation parameter sets, as well as multiple static evaluation parameters, is calculated to obtain multiple dynamic mutual exclusion coefficients. This process calculates the difference between the dynamic evaluation parameter set of each pipeline segment and the dynamic evaluation parameter sets of other pipeline segments, as well as the difference between the dynamic evaluation parameter set and the static evaluation parameters. Specifically, for the i-th pipeline segment, the deviation of its dynamic evaluation parameter set from the dynamic evaluation parameter sets of the other n-1 pipeline segments is calculated, along with the deviation from the static evaluation parameters. These deviation values ​​are then statistically averaged to obtain the dynamic mutual exclusion coefficient corresponding to the i-th pipeline segment. Since there are multiple pipeline segments, multiple dynamic mutual exclusion coefficients are obtained, each reflecting the degree of difference between the monitoring data of the corresponding pipeline segment and other data sources.

[0050] Through the above mutual exclusion analysis calculations, static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients were obtained. These coefficients characterize the deviation level between different evaluation parameter data sources, providing a data quality assessment basis for subsequent weight allocation and control scheme optimization in the VIKOR algorithm, and ensuring the reliability of risk assessment results and the rationality of control schemes.

[0051] Furthermore, the difference between the multiple static evaluation parameters and the multiple dynamic evaluation parameter sets is calculated to obtain the static mutual exclusion coefficient, including:

[0052] S211. Calculate the difference magnitude between the plurality of static evaluation parameters and the plurality of dynamic evaluation parameters in each dynamic evaluation parameter set to obtain a plurality of static difference magnitude sets;

[0053] S212. Assign weights to multiple evaluation indicators, and perform weighted calculations on multiple static difference amplitudes within multiple static difference amplitude sets to obtain static mutual exclusion coefficients.

[0054] In a preferred embodiment, firstly, the difference magnitude between multiple static evaluation parameters and multiple dynamic evaluation parameters within each dynamic evaluation parameter set is calculated to obtain multiple sets of static difference magnitudes. This process calculates the difference magnitude between the dynamic evaluation parameter set and the static evaluation parameters for each pipeline segment. Specifically, for the dynamic evaluation parameter set of the i-th pipeline segment, the absolute value difference between the dynamic evaluation parameters of each evaluation index of that pipeline segment and the corresponding static evaluation parameters is calculated one by one to obtain the set of static difference magnitudes for that pipeline segment. For example, for the first pipeline segment, the difference magnitude between its operating pressure, wall thickness, corrosion potential, and other evaluation indices and the static evaluation parameters is calculated to form the first set of static difference magnitudes. Similarly, for each of the other pipeline segments, the corresponding set of static difference magnitudes is calculated to obtain the final set of static difference magnitudes, resulting in multiple sets of static difference magnitudes equal to the number of pipeline segments.

[0055] Then, weights are assigned to multiple evaluation indicators, and the weighted calculations are performed on multiple static difference amplitudes within multiple static difference amplitude sets to obtain the static mutual exclusion coefficient. This process first assigns corresponding indicator weights based on the importance of each evaluation indicator in the risk assessment. These evaluation indicators include pipeline body indicators such as pipeline material, wall thickness, welding quality, and anti-corrosion coating condition; environmental factor indicators such as soil corrosivity, geological stability, and third-party construction activities; and operational status indicators such as operating pressure, flow rate changes, and temperature distribution. Weights are assigned to each evaluation indicator based on expert experience; for example, operating pressure is weighted at 0.15, wall thickness at 0.12, and anti-corrosion coating condition at 0.10, ensuring that the sum of all evaluation indicator weights is 1. Then, for each static difference amplitude set, each difference amplitude within the set is multiplied by its corresponding indicator weight, and the sum is obtained to obtain the weighted difference degree for that pipeline segment. Finally, the weighted difference degrees of all pipeline segments are averaged to finally obtain the static mutual exclusion coefficient. This static mutual exclusion coefficient comprehensively considers the differences in importance of different evaluation indicators and accurately quantifies the overall deviation level between expert assessment data and real-time monitoring data.

[0056] Furthermore, risk analysis is conducted based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results, including:

[0057] S31. Obtain the risk analyzer;

[0058] S32. Input the multiple static evaluation parameters and multiple dynamic evaluation parameter sets into the risk analyzer respectively, and output the static risk parameters and multiple dynamic risk parameters as the risk assessment results.

[0059] In a preferred embodiment, firstly, a risk analyzer is acquired. The risk analyzer is a risk assessment model built using machine learning techniques, used to convert input evaluation parameters into quantified risk parameters. The risk analyzer includes a static risk analysis branch and a dynamic risk analysis branch, respectively handling multiple sets of static evaluation parameters and multiple sets of dynamic evaluation parameters. The static risk analysis branch specifically receives and processes static evaluation parameters, learning the mapping relationship between static evaluation parameters and risk levels through training; the dynamic risk analysis branch specifically receives and processes sets of dynamic evaluation parameters, learning the correlation between real-time monitoring data and risk status through training. The risk analyzer is constructed based on historical oil and gas pipeline risk management data, and is trained using deep neural networks or other machine learning algorithms to ensure accurate identification of risk parameters corresponding to different combinations of evaluation parameters.

[0060] Subsequently, the multiple static evaluation parameters and multiple dynamic evaluation parameter sets are input into the risk analyzer, which outputs static risk parameters and multiple dynamic risk parameters as the risk assessment results. In specific implementation, the static risk analysis branch receives multiple static evaluation parameters, including pipeline material, wall thickness, operating pressure, and soil corrosivity, as input, and outputs static risk parameters that reflect the inherent risk level of the pipeline based on expert assessment. Simultaneously, the dynamic risk analysis branch receives the dynamic evaluation parameter sets of each pipeline segment as input, independently analyzes the real-time monitoring data of each pipeline segment, and outputs corresponding dynamic risk parameters. These dynamic risk parameters reflect the current real-time risk status of each pipeline segment.

[0061] Through the above risk analysis process, risk assessment results covering both static and dynamic dimensions were obtained, providing a quantitative basis for subsequent optimization of control measures.

[0062] Furthermore, acquire risk analyzers, including:

[0063] S311. Based on the risk management data of oil and gas pipelines over a historical period, collect a set of static evaluation parameter groups for samples and a set of dynamic evaluation parameter groups for multiple samples, and collect a set of static risk parameters and a set of dynamic risk parameters for samples.

[0064] S312. Based on machine learning, a risk analyzer is constructed, wherein the risk analyzer includes a static risk analysis branch and a dynamic risk analysis branch;

[0065] S313. Using the set of static evaluation parameters and the set of static risk parameters of the samples, supervised training and testing are performed on the static risk analysis branch until convergence is achieved. Using the set of multiple dynamic evaluation parameters and the set of dynamic risk parameters of the samples, supervised training and testing are performed on the dynamic risk analysis branch until convergence is achieved.

[0066] S314. Based on the convergent static risk analysis branch and dynamic risk analysis branch, obtain the risk analyzer.

[0067] In a preferred embodiment, firstly, based on historical risk management data of oil and gas pipelines, a set of sample static evaluation parameter sets and multiple sets of sample dynamic evaluation parameter sets are collected, along with a set of sample static risk parameters and a set of sample dynamic risk parameters. Specifically, a large amount of labeled risk management data is extracted from the historical database, including expert evaluation records and sensor monitoring records for different periods and pipelines. The sample static evaluation parameter set contains expert evaluation data on various indicators such as pipeline material, wall thickness, and environmental factors from historical cases; the multiple sets of sample dynamic evaluation parameter sets contain real-time operational data collected by sensors on various pipeline sections from historical cases; the sample static risk parameter set contains the corresponding expert risk evaluation results, such as risk level and safety evaluation conclusions; and the sample dynamic risk parameter set contains corresponding records of actual risk events, such as leaks and equipment failures. This sample data constitutes the basic dataset for machine learning training, ensuring that the model can learn the inherent correlation between evaluation parameters and risk results.

[0068] Then, a risk analyzer is constructed based on machine learning, comprising a static risk analysis branch and a dynamic risk analysis branch. The static risk analysis branch can employ machine learning algorithms such as fully connected neural networks, support vector machines, and random forests to learn the mapping relationship between static evaluation parameters and static risk parameters. The dynamic risk analysis branch can employ algorithms suitable for processing time-series data, such as recurrent neural networks, long short-term memory networks, and convolutional neural networks, to learn the correlation between the temporal change patterns of dynamic monitoring data and dynamic risk parameters. The two branches are structurally independent, but information can be fused at the output layer to form a complete risk analyzer architecture. For example, suppose multiple evaluation indicators include 15 indicators: pipe material, wall thickness, welding quality, anti-corrosion coating condition, service life, soil corrosivity, geological stability, third-party construction activities, population density, ambient temperature, operating pressure, flow rate change, temperature distribution, vibration frequency, and pressure fluctuation amplitude. The static risk analysis branch employs a three-layer fully connected neural network structure. The input layer contains 15 neurons corresponding to the aforementioned 15 static evaluation indicators, the hidden layer contains 32 neurons using the ReLU activation function, and the output layer contains one neuron that outputs a static risk score of 0-100, serving as the static risk parameter. The dynamic risk analysis branch employs an LSTM recurrent neural network structure. The input layer receives 24 hours of time-series data, with each time step containing the aforementioned 15 evaluation indicators. The LSTM hidden layer contains 64 units, and the output layer is mapped to a dynamic risk score of 0-100 through a fully connected layer, serving as the dynamic risk parameter.

[0069] Next, supervised training was performed on both the static and dynamic risk analysis branches. For example, for the static risk analysis branch, a set of static evaluation parameters was used as input. Each sample contained a 15-dimensional feature vector, including pipe material grade, wall thickness, and soil corrosivity level. The corresponding static risk parameter set was a risk score of 0-100 as assessed by experts, serving as the supervision label. The training process used the Adam optimizer with a learning rate of 0.001 and a batch size of 64. The error between the predicted and actual scores was calculated using the mean squared error loss function. After multiple rounds of training, the training converged, resulting in a converged static risk analysis branch. Similarly, for the dynamic risk analysis branch, multiple sets of dynamic evaluation parameters were used. Each sample was a 24-hour × 15-dimensional time-series matrix containing real-time monitoring data such as operating pressure, flow rate changes, and temperature distribution. The corresponding dynamic risk parameter set was the risk score. The training process employed the same optimization strategy, and after multiple rounds of training, the training converged, resulting in a converged dynamic risk analysis branch.

[0070] Subsequently, based on the converged static and dynamic risk analysis branches, a complete risk analyzer is obtained. The two converged branches possess the ability to accurately predict the risks of corresponding types of evaluation parameters. Through model ensemble, they are combined into a unified risk analyzer. This risk analyzer can simultaneously process both static and dynamic evaluation parameter sets, outputting corresponding risk assessment results, providing reliable technical support for oil and gas pipeline risk management.

[0071] Furthermore, based on static risk parameters and multiple dynamic risk parameters, as well as static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, the VIKOR algorithm is used to optimize the control scheme and obtain the optimal control scheme, including:

[0072] S41. Obtain the control scheme space for oil and gas pipeline risk management and randomly generate the first control scheme;

[0073] S42. Analyze the first control scheme with the first static control score and the first dynamic control score of the static risk parameters and multiple dynamic risk parameters;

[0074] S43. Based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, the first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and the first control value is also calculated.

[0075] S44. Continue to optimize the control plan to obtain the optimal control plan with the maximum control value.

[0076] In a preferred embodiment, firstly, a control scheme space for oil and gas pipeline risk management is obtained, and a first control scheme is randomly generated. The control scheme space contains multiple possible combinations of risk management strategies, covering various control dimensions such as maintenance plans, inspection frequency adjustments, pressure control measures, emergency response plans, and equipment replacement plans. Each control scheme consists of multiple control parameters, such as maintenance cycle, inspection interval, pressure threshold setting, emergency response time, and equipment maintenance level. Values ​​of various control parameters are randomly selected from the control scheme space and combined to generate the first control scheme as an initial candidate solution for the VIKOR algorithm, providing a starting point for subsequent iterative optimization.

[0077] Then, the first control scheme is analyzed in conjunction with static risk parameters, multiple dynamic risk parameters, and first static control scores, as well as multiple first dynamic control scores. Specifically, this is achieved through a pre-trained control predictor, which is a predictive model built based on historical control scheme implementation effect data. The first control scheme is combined with static risk parameters and input into the control predictor, outputting a first static control score, which reflects the expected effect of the first control scheme under the current static risk conditions. Simultaneously, the first control scheme is combined with the dynamic risk parameters of each pipeline segment, outputting multiple first dynamic control scores, which respectively reflect the expected effect of the first control scheme under the current dynamic risk conditions of each pipeline segment.

[0078] Subsequently, based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, a first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and then to calculate the first control value. The VIKOR algorithm first uses the static mutual exclusion coefficients as static regret weights to calculate the corresponding static utility weights, and uses the multiple dynamic mutual exclusion coefficients as multiple dynamic regret weights to calculate the corresponding dynamic utility weights. Then, it calculates the difference between each control score and the maximum control score, and combines this with the corresponding regret weights to calculate multiple candidate individual regret values; based on each dynamic control score and its corresponding utility weight, it calculates multiple candidate group utility values. Next, it selects the largest candidate group utility value and its corresponding candidate individual regret value as the first group utility value and the first individual regret value. Finally, based on the fused regret weight and fused utility weight, it performs a weighted calculation on the first group utility value and the first individual regret value to obtain the first control value.

[0079] Subsequently, iterative optimization of the control scheme continues. New candidate control schemes are generated within the control scheme space, and steps S42 and S43 are repeated to obtain the corresponding control values. For example, optimization algorithms such as genetic algorithms, particle swarm optimization, or simulated annealing can be used to intelligently search within the control scheme space, continuously generating candidate solutions with better performance. After multiple iterative optimizations, the optimal control scheme with the maximum control value is finally obtained. This optimal control scheme achieves the best balance between improving overall control effectiveness and reducing individual shortcomings, providing an effective risk management strategy for the target oil and gas pipeline.

[0080] Furthermore, the analysis includes the first control scheme and the first static control score and multiple first dynamic control scores of the static risk parameters and multiple dynamic risk parameters, including:

[0081] S421. Obtain the control predictor, wherein the control predictor is obtained by training and convergence using a set of sample control schemes, a set of sample risk parameter groups, and a set of sample control scores.

[0082] S422. The first control scheme is combined with static risk parameters and multiple dynamic risk parameters and input into the control predictor, and the first static control score and multiple first dynamic control scores are output respectively.

[0083] In a preferred embodiment, firstly, a control prediction engine is obtained. This engine is trained and converged using a set of sample control schemes, a set of sample risk parameters, and a set of sample control scores. It is used to predict the effectiveness of a specific control scheme under given risk conditions. The set of sample control schemes includes various combinations of historically implemented control strategies, such as different maintenance cycles, inspection frequencies, and pressure control parameters. The set of sample risk parameters includes static and dynamic risk scores for the corresponding period. The set of sample control scores includes evaluations of the actual control implementation results, such as quantitative indicators like risk reduction, cost-effectiveness, and implementation feasibility. The control prediction engine employs a supervised learning method, using the combination of control schemes and risk parameters as input features and the control scores as the output target. It is trained using algorithms such as neural networks, support vector machines, or ensemble learning until it converges and possesses accurate predictive capabilities.

[0084] Subsequently, the first control scheme is combined with static risk parameters and multiple dynamic risk parameters and input into the control predictor, respectively, to obtain a first static control score and multiple first dynamic control scores. Specifically, firstly, the parameters of the first control scheme are combined with the static risk parameters to form a static control input vector. This vector is input into the control predictor, which outputs a first static control score based on learned historical experience patterns. This first static control score reflects the expected control effect of the first control scheme under the current static risk conditions. Simultaneously, the first control scheme is combined with the dynamic risk parameters of each pipeline segment to form multiple dynamic control input vectors, which are then input into the control predictor. The model outputs multiple first dynamic control scores, which reflect the expected control effect of the control scheme under the current dynamic risk conditions of each pipeline segment. Through the above prediction process, quantitative evaluation results of the first control scheme under different risk scenarios are obtained, providing basic data for the subsequent calculation of the group utility and individual regret of the VIKOR algorithm.

[0085] Furthermore, based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, a first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and to calculate the first control value, including:

[0086] S431. Use the static mutual exclusion coefficient as the static regret weight, and calculate the static utility weight;

[0087] S432. Use multiple dynamic mutual exclusion coefficients as multiple dynamic regret weights, and calculate multiple dynamic utility weights;

[0088] S433. Calculate the difference between the first static control score and the maximum control score, and combine it with the static regret weight to calculate the regret value of the first candidate individual.

[0089] S434. Calculate the utility value of the first candidate group based on multiple dynamic control scores and multiple dynamic utility weights.

[0090] S435. Continue to calculate the regret values ​​of multiple first candidate individuals and the utility values ​​of the first candidate group, and select the largest first candidate group utility value and the corresponding first candidate individual regret value as the first group utility value and the first individual regret value;

[0091] S436. Based on the static mutual exclusion coefficient and multiple dynamic mutual exclusion coefficients, calculate the fusion regret weight and the fusion utility weight;

[0092] S437. Based on the fusion utility weight and fusion regret weight, the first group utility value and the first individual regret value are weighted and calculated to obtain the first control value, wherein the weighted first group utility value is subtracted from the weighted first individual regret value.

[0093] In a preferred embodiment, firstly, the static mutual exclusion coefficient is used as the static regret weight, and the static utility weight is calculated. A larger mutual exclusion coefficient indicates a greater difference from other data sources, potentially leading to lower data reliability; therefore, a larger regret weight is set to reduce its impact. Since the mutual exclusion coefficient itself is already a value within the range of 0-1, the static mutual exclusion coefficient is directly assigned as the static regret weight. Then, the static utility weight is calculated using the formula: Static Utility Weight = 1 - Static Regret Weight, ensuring that the sum of the two weights is 1, thus achieving a reasonable allocation of weights.

[0094] Then, multiple dynamic mutual exclusion coefficients are used as multiple dynamic regret weights, and multiple dynamic utility weights are calculated. For each pipe segment, the corresponding dynamic mutual exclusion coefficient is assigned a dynamic regret weight, and then the dynamic utility weight corresponding to each pipe segment is calculated using the formula: Dynamic Utility Weight = 1 - Dynamic Regret Weight, forming multiple dynamic utility weights equal to the number of pipe segments.

[0095] Subsequently, the difference between the first static control score and the maximum control score is calculated, and combined with the static regret weight, the regret value of the first first candidate individual is calculated. First, the maximum value among the first static control score and multiple first dynamic control scores is identified as the maximum control score. Then, the difference between the first static control score and the maximum control score is calculated, and this difference is multiplied by the static regret weight to obtain the regret value of the first first candidate individual. Simultaneously, based on multiple dynamic control scores and multiple dynamic utility weights, the utility value of the first first candidate group is calculated. Each of the multiple first dynamic control scores is multiplied by its corresponding dynamic utility weight, and then all products are summed to obtain the utility value of the first first candidate group. This value reflects the weighted comprehensive performance of the control effectiveness of each pipeline segment.

[0096] Next, the regret values ​​of multiple first-candidate individuals and the utility values ​​of the first-candidate group are calculated. For each first dynamic control score, the difference between it and the maximum control score is calculated, and multiplied by the corresponding dynamic regret weight to obtain multiple first-candidate individual regret values. Simultaneously, using each dynamic control score as the primary factor, combined with other control scores and utility weights, multiple first-candidate group utility values ​​are calculated. Finally, the largest first-candidate group utility value and its corresponding first-candidate individual regret value among all candidates are selected as the first group utility value and the first individual regret value.

[0097] Then, based on the static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, the fusion regret weight and the fusion utility weight are calculated. Specifically, the fusion regret weight is calculated by taking the arithmetic mean of the static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, and then the fusion utility weight is calculated using the formula: Fusion Utility Weight = 1 - Fusion Regret Weight.

[0098] Finally, based on the fusion utility weight and fusion regret weight, the first group utility value and the first individual regret value are weighted and calculated to obtain the first control value. The formula is: First Control Value = Fusion Utility Weight × First Group Utility Value - Fusion Regret Weight × First Individual Regret Value. By subtracting the weighted individual regret from the weighted group utility, the first control value, which comprehensively balances group interests and individual losses, is obtained. A larger first control value indicates a better control plan.

[0099] Example 2, as Figure 2 As shown, based on the same inventive concept as the oil and gas pipeline risk assessment method integrating static and dynamic analysis provided in Embodiment 1, this embodiment of the invention also provides an oil and gas pipeline risk assessment system integrating static and dynamic analysis, including:

[0100] The data acquisition module 11 is used to acquire multiple evaluation indicators for risk assessment of the target oil and gas pipeline, acquire multiple static evaluation parameters of the multiple evaluation indicators, and collect multiple sets of dynamic evaluation parameters.

[0101] The mutual exclusion analysis module 12 is used to perform mutual exclusion analysis on the multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients.

[0102] Risk analysis module 13 is used to perform risk analysis based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results;

[0103] The scheme optimization module 14 is used to optimize the control scheme using the VIKOR algorithm based on static risk parameters and multiple dynamic risk parameters, as well as static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, to obtain the optimal control scheme and carry out risk control.

[0104] Furthermore, the execution steps of the data acquisition module 11 include:

[0105] Obtain multiple evaluation indicators for risk assessment of oil and gas pipelines;

[0106] Multiple static evaluation parameters of the target oil and gas pipeline were collected according to multiple evaluation indicators.

[0107] The target oil and gas pipeline is divided into multiple pipeline segments. Through sensors, dynamic evaluation parameters of multiple evaluation indicators are collected to obtain multiple sets of dynamic evaluation parameters.

[0108] Furthermore, the execution steps of the mutual exclusion analysis module 12 include:

[0109] Calculate the mean difference between the multiple static evaluation parameters and the multiple dynamic evaluation parameter sets to obtain the static mutual exclusion coefficient;

[0110] Calculate the mean difference between each dynamic evaluation parameter set and other dynamic evaluation parameter sets, as well as multiple static evaluation parameters, to obtain multiple dynamic mutual exclusion coefficients.

[0111] Furthermore, the execution steps of the mutual exclusion analysis module 12 also include:

[0112] Calculate the difference magnitude between the plurality of static evaluation parameters and the plurality of dynamic evaluation parameters in each dynamic evaluation parameter set to obtain a plurality of static difference magnitude sets;

[0113] Assign weights to multiple evaluation indicators, and perform weighted calculations on multiple static difference amplitudes within multiple static difference amplitude sets to obtain static mutual exclusion coefficients.

[0114] Furthermore, the execution steps of risk analysis module 13 include:

[0115] Obtain the risk analyzer;

[0116] The multiple static evaluation parameters and multiple dynamic evaluation parameter sets are respectively input into the risk analyzer, and the static risk parameters and multiple dynamic risk parameters are output as the risk assessment results.

[0117] Furthermore, the execution steps of risk analysis module 13 also include:

[0118] Based on historical risk management data of oil and gas pipelines, a set of static evaluation parameter groups and a set of dynamic evaluation parameter groups for multiple samples were collected, as well as a set of static risk parameters and a set of dynamic risk parameters for samples.

[0119] A risk analyzer is constructed based on machine learning, wherein the risk analyzer includes a static risk analysis branch and a dynamic risk analysis branch;

[0120] The static risk analysis branch is trained and tested under supervision using the set of static evaluation parameters and the set of static risk parameters for the samples until convergence. The dynamic risk analysis branch is trained and tested under supervision using the set of multiple dynamic evaluation parameters and the set of dynamic risk parameters for the samples until convergence.

[0121] A risk analyzer is obtained based on the convergent static risk analysis branch and the dynamic risk analysis branch.

[0122] Furthermore, the execution steps of the scheme optimization module 14 include:

[0123] Obtain the control scheme space for oil and gas pipeline risk management, and randomly generate the first control scheme;

[0124] Analyze the first control scheme with the first static control score and the first dynamic control score of the static risk parameters and multiple dynamic risk parameters;

[0125] Based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, the first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and to calculate the first control value.

[0126] Continue to optimize the control plan to obtain the optimal control plan with the maximum control value.

[0127] Furthermore, the execution steps of the scheme optimization module 14 also include:

[0128] A control prediction tool is obtained, wherein the control prediction tool is trained and converged using a set of sample control schemes, a set of sample risk parameter groups, and a set of sample control scores;

[0129] The first control scheme is combined with static risk parameters and multiple dynamic risk parameters and input into the control predictor, and the first static control score and multiple first dynamic control scores are output respectively.

[0130] Furthermore, the execution steps of the scheme optimization module 14 also include:

[0131] The static mutual exclusion coefficient is used as the static regret weight, and the static utility weight is calculated.

[0132] Multiple dynamic mutual exclusion coefficients are used as multiple dynamic regret weights, and multiple dynamic utility weights are calculated.

[0133] Calculate the difference between the first static control score and the maximum control score, and combine it with the static regret weight to calculate the regret value of the first candidate individual;

[0134] Calculate the utility value of the first candidate group based on multiple dynamic control scores and multiple dynamic utility weights.

[0135] Continue to calculate the regret values ​​of multiple first candidate individuals and the utility values ​​of the first candidate group, and select the largest first candidate group utility value and the corresponding first candidate individual regret value as the first group utility value and the first individual regret value;

[0136] Based on the static mutual exclusion coefficient and multiple dynamic mutual exclusion coefficients, calculate the fusion regret weight and the fusion utility weight;

[0137] Based on the fusion utility weight and fusion regret weight, the first group utility value and the first individual regret value are weighted and calculated to obtain the first control value, wherein the weighted first group utility value is subtracted from the weighted first individual regret value.

[0138] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0139] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0140] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0141] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0143] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0144] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A risk assessment method for oil and gas pipelines that integrates static and dynamic analysis, characterized in that, The method includes: Multiple evaluation indicators for risk assessment of target oil and gas pipelines are obtained, along with multiple static evaluation parameters for these indicators and multiple sets of dynamic evaluation parameters. Mutual exclusion analysis is performed on the multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients; Risk analysis is conducted based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results; Based on static risk parameters and multiple dynamic risk parameters, as well as static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, the VIKOR algorithm is used to optimize the control scheme, obtain the optimal control scheme, and perform risk control, including: Obtain the control scheme space for oil and gas pipeline risk management, and randomly generate the first control scheme; Analyze the first control scheme with the first static control score and the first dynamic control score of the static risk parameters and multiple dynamic risk parameters; Based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, a first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and to calculate the first control value, including: The static mutual exclusion coefficient is used as the static regret weight, and the static utility weight is calculated. Multiple dynamic mutual exclusion coefficients are used as multiple dynamic regret weights, and multiple dynamic utility weights are calculated. Calculate the difference between the first static control score and the maximum control score, and combine it with the static regret weight to calculate the regret value of the first candidate individual; Calculate the utility value of the first candidate group based on multiple dynamic control scores and multiple dynamic utility weights. Continue to calculate the regret values ​​of multiple first candidate individuals and the utility values ​​of the first candidate group, and select the largest first candidate group utility value and the corresponding first candidate individual regret value as the first group utility value and the first individual regret value; Based on the static mutual exclusion coefficient and multiple dynamic mutual exclusion coefficients, calculate the fusion regret weight and the fusion utility weight; Based on the fusion utility weight and fusion regret weight, the first group utility value and the first individual regret value are weighted and calculated to obtain the first control value, wherein the weighted first group utility value is subtracted from the weighted first individual regret value. Continue to optimize the control plan to obtain the optimal control plan with the maximum control value.

2. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 1, characterized in that, Multiple evaluation indicators for risk assessment of the target oil and gas pipeline are obtained, along with multiple static evaluation parameters for these indicators, and multiple sets of dynamic evaluation parameters are collected, including: Obtain multiple evaluation indicators for risk assessment of oil and gas pipelines; Multiple static evaluation parameters of the target oil and gas pipeline were collected according to multiple evaluation indicators. The target oil and gas pipeline is divided into multiple pipeline segments. Through sensors, dynamic evaluation parameters of multiple evaluation indicators are collected to obtain multiple sets of dynamic evaluation parameters.

3. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 1, characterized in that, Mutual exclusion analysis is performed on the multiple sets of static evaluation parameters and multiple sets of dynamic evaluation parameters to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients, including: Calculate the mean difference between the multiple static evaluation parameters and the multiple dynamic evaluation parameter sets to obtain the static mutual exclusion coefficient; Calculate the mean difference between each dynamic evaluation parameter set and other dynamic evaluation parameter sets, as well as multiple static evaluation parameters, to obtain multiple dynamic mutual exclusion coefficients.

4. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 3, characterized in that, Calculate the difference between the multiple static evaluation parameters and the multiple dynamic evaluation parameter sets to obtain the static mutual exclusion coefficient, including: Calculate the difference magnitude between the plurality of static evaluation parameters and the plurality of dynamic evaluation parameters in each dynamic evaluation parameter set to obtain a plurality of static difference magnitude sets; Assign weights to multiple evaluation indicators, and perform weighted calculations on multiple static difference amplitudes within multiple static difference amplitude sets to obtain static mutual exclusion coefficients.

5. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 1, characterized in that, Based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets, risk analysis is performed separately to obtain static risk parameters and multiple dynamic risk parameters, which serve as the risk assessment results, including: Obtain the risk analyzer; The multiple static evaluation parameters and multiple dynamic evaluation parameter sets are respectively input into the risk analyzer, and the static risk parameters and multiple dynamic risk parameters are output as the risk assessment results.

6. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 5, characterized in that, Obtain risk analyzers, including: Based on historical risk management data of oil and gas pipelines, a set of static evaluation parameter groups and a set of dynamic evaluation parameter groups for multiple samples were collected, as well as a set of static risk parameters and a set of dynamic risk parameters for samples. A risk analyzer is constructed based on machine learning, wherein the risk analyzer includes a static risk analysis branch and a dynamic risk analysis branch; The static risk analysis branch is trained and tested under supervision using the set of static evaluation parameters and the set of static risk parameters for the samples until convergence. The dynamic risk analysis branch is trained and tested under supervision using the set of multiple dynamic evaluation parameters and the set of dynamic risk parameters for the samples until convergence. A risk analyzer is obtained based on the convergent static risk analysis branch and the dynamic risk analysis branch.

7. The oil and gas pipeline risk assessment method integrating static and dynamic analysis according to claim 1, characterized in that, The analysis of the first control scheme and the first static control score and multiple first dynamic control scores of the static risk parameters and multiple dynamic risk parameters includes: A control prediction tool is obtained, wherein the control prediction tool is trained and converged using a set of sample control schemes, a set of sample risk parameter groups, and a set of sample control scores; The first control scheme is combined with static risk parameters and multiple dynamic risk parameters and input into the control predictor, and the first static control score and multiple first dynamic control scores are output respectively.

8. A risk assessment system for oil and gas pipelines integrating static and dynamic analysis, characterized in that, The system is used to implement the oil and gas pipeline risk assessment method integrating static and dynamic analysis as described in any one of claims 1 to 7, the system comprising: The data acquisition module is used to acquire multiple evaluation indicators for risk assessment of the target oil and gas pipeline, acquire multiple static evaluation parameters for multiple evaluation indicators, and collect multiple sets of dynamic evaluation parameters. The mutual exclusion analysis module is used to perform mutual exclusion analysis on the multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static mutual exclusion coefficients and multiple dynamic mutual exclusion coefficients. The risk analysis module is used to perform risk analysis based on multiple static evaluation parameters and multiple dynamic evaluation parameter sets to obtain static risk parameters and multiple dynamic risk parameters as risk assessment results. The scheme optimization module is used to optimize the control scheme based on static risk parameters, multiple dynamic risk parameters, static mutual exclusion coefficients, and multiple dynamic mutual exclusion coefficients, using the VIKOR algorithm to obtain the optimal control scheme for risk management. This includes: Obtain the control scheme space for oil and gas pipeline risk management, and randomly generate the first control scheme; Analyze the first control scheme with the first static control score and the first dynamic control score of the static risk parameters and multiple dynamic risk parameters; Based on static mutual exclusion coefficients, multiple dynamic mutual exclusion coefficients, a first static control score, and multiple first dynamic control scores, the VIKOR algorithm is used to calculate the first group utility value and the first individual regret value, and to calculate the first control value, including: The static mutual exclusion coefficient is used as the static regret weight, and the static utility weight is calculated. Multiple dynamic mutual exclusion coefficients are used as multiple dynamic regret weights, and multiple dynamic utility weights are calculated. Calculate the difference between the first static control score and the maximum control score, and combine it with the static regret weight to calculate the regret value of the first candidate individual; Calculate the utility value of the first candidate group based on multiple dynamic control scores and multiple dynamic utility weights. Continue to calculate the regret values ​​of multiple first candidate individuals and the utility values ​​of the first candidate group, and select the largest first candidate group utility value and the corresponding first candidate individual regret value as the first group utility value and the first individual regret value; Based on the static mutual exclusion coefficient and multiple dynamic mutual exclusion coefficients, calculate the fusion regret weight and the fusion utility weight; Based on the fusion utility weight and fusion regret weight, the first group utility value and the first individual regret value are weighted and calculated to obtain the first control value, wherein the weighted first group utility value is subtracted from the weighted first individual regret value. Continue to optimize the control plan to obtain the optimal control plan with the maximum control value.

Images