Buried natural gas gathering pipeline leakage prediction method and model

Abstract

The application relates to a buried natural gas gathering pipeline leakage prediction method and model, and belongs to the technical field of natural gas new energy storage and production safety detection, and particularly relates to a buried natural gas gathering pipeline leakage prediction method and model, and the specific method is as follows: a leakage model with the same pipe diameter, different materials, buried depths, leakage holes and temperatures is established according to the same proportion principle by using finite elements in combination with the service environment and working conditions of the buried natural gas pipeline site; the leakage model transportation medium proportion is set through a multi-component material parameter setting module, and different working conditions are simulated to establish a buried natural gas gathering pipeline leakage simulation calculation model with different physical properties; the application can accurately, quickly and efficiently find the leakage point and leakage condition, and can dynamically monitor whether the pipeline has a leakage point in real time during the operation of the buried natural gas gathering pipeline, so that explosion accidents caused by the leakage of the buried natural gas gathering pipeline can be effectively avoided.

Description

Technical Field

[0001] This invention relates to the field of natural gas new energy storage, transportation and production safety detection technology, specifically to a method and model for predicting leaks in buried natural gas gathering and transmission pipelines. Background Technology

[0002] Buried natural gas gathering and transmission pipelines are highly susceptible to leakage and damage under both external and internal corrosion, posing a significant threat to the production, storage, and transportation of energy from these pipelines. Under external conditions such as soil, rainwater, and temperature, the surface of buried natural gas gathering and transmission pipelines is prone to electrochemical corrosion and rust, leading to localized corrosion leaks. Internally, the transported medium causes varying degrees of chemical corrosion on the inner wall surface of the pipelines, reducing the strength of the pipeline materials. In severe cases, this can lead to localized cracks or damage, ultimately causing leaks of the transported substances.

[0003] Buried natural gas gathering and transmission pipelines can also experience leaks due to factors such as manufacturer production, manual installation, and service environment. Because of the concealed nature of being buried, leaks in the early stages are extremely difficult to detect and discover. This is a significant factor leading to large-scale accidents caused by leaks in buried natural gas gathering and transmission pipelines. Therefore, it is necessary to explore finite element model calculations and big data prediction models for leaks in buried natural gas gathering and transmission pipelines. Summary of the Invention

[0004] The purpose of this invention is to provide a method and model for predicting leaks in buried natural gas gathering and transmission pipelines, which solves the technical problem that existing leak detection technologies and methods for buried natural gas gathering and transmission pipelines cannot accurately, quickly, and efficiently detect leak points and leak conditions, and are difficult to monitor the pipeline for leaks in real time during operation, thus failing to effectively avoid explosion accidents caused by leaks in buried natural gas gathering and transmission pipelines.

[0005] This invention discloses a method for predicting leakage in buried natural gas gathering and transmission pipelines, the specific method of which is as follows:

[0006] Using the finite element method combined with the service environment and working conditions of the buried natural gas pipeline, a leakage model with the same pipe diameter, different materials, burial depth, leakage hole, and temperature was established according to the principle of proportionality.

[0007] The proportion of the transport medium in the leakage model is set by the multi-component material parameter setting module, and different working conditions are simulated to establish a leakage simulation calculation model for buried natural gas gathering and transmission pipelines with different physical properties. Based on the leakage simulation calculation model of buried natural gas gathering and transmission pipelines with different physical properties, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are simulated and calculated to obtain the data of the pipeline leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point over time. A set of the above data is obtained at the required time interval to obtain the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties.

[0008] The simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties are cleaned, transformed and standardized by multi-source heterogeneous data fusion technology to obtain the consistent change relationship between leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point and time. The change characteristics of each physical quantity of buried natural gas gathering and transmission pipeline leakage are analyzed, and corresponding prediction conclusions of buried natural gas gathering and transmission pipeline leakage are given.

[0009] Based on the prediction conclusions corresponding to the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties, a training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model is generated, which includes leakage amount, leakage velocity, pressure before the leakage point, and pressure drop near the leakage point.

[0010] Using a multinomial linear regression model in machine learning, a training sample dataset for a leak prediction model of buried natural gas gathering and transmission pipelines was used to train and optimize the leak model, resulting in an optimized training sample dataset for the leak prediction model. The linear error of the training sample dataset was analyzed, and the training sample dataset for the pipeline leak prediction model regarding leakage amount, leakage velocity, pressure before the leak point, and pressure drop near the leak point was obtained, thus establishing a preliminary leak prediction model for buried natural gas gathering and transmission pipelines.

[0011] By conducting indoor leakage characteristic tests on buried natural gas gathering and transmission pipelines, the correctness and effectiveness of the preliminary leakage prediction model for buried natural gas gathering and transmission pipelines were verified. The characteristics of leakage results and the differences in prediction results under the same test conditions were analyzed. Error analysis, a mathematical analysis method, was used to analyze the leakage prediction error of buried natural gas gathering and transmission pipelines, and the preliminary leakage prediction model was optimized. The resulting buried natural gas gathering and transmission pipeline leakage prediction model is then obtained. When leakage prediction is needed, the operating parameters of the buried natural gas pipeline are input into the prediction model. The model is then compared and analyzed with the training sample dataset of the pipeline leakage prediction model to determine whether a leak has occurred in the pipeline.

[0012] Preferably, when establishing a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the Fluent fluid field module in ANSYS Workbench is selected as the modeling and simulation environment. The Mixture Material multi-component material parameter setting module is used to set the proportions of CH4, H2S, H2O, CO2, O2, C2H6, H2, and N2 components in the natural gas pipeline. According to the actual size and design specifications of the buried natural gas gathering and transmission pipeline, a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties is established.

[0013] Preferably, the data on the changes in pipeline leakage rate, leakage velocity, pressure upstream of the leakage point, and pressure drop near the leakage point over time are obtained in the following manner:

[0014] Based on the establishment of a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the inlet pressure and velocity parameters of multi-component gas are set at the front end of the natural gas pipeline, and the outlet pressure and velocity parameters of multi-component gas are set at the rear end of the natural gas pipeline. The porosity, viscous resistance coefficient and inertial resistance coefficient of the soil outside the buried natural gas gathering and transmission pipeline are set. Through simulation calculation, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are obtained from the simulation calculation model of leakage of buried natural gas gathering and transmission pipeline over time.

[0015] Preferably, the training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model includes training sample datasets of leakage amount versus time, leakage velocity versus time, pressure versus time before the leakage point, and pressure drop versus time near the leakage point.

[0016] A leakage prediction model for buried natural gas gathering and transmission pipelines is provided, which is obtained using one of the aforementioned leakage prediction methods for buried natural gas gathering and transmission pipelines.

[0017] This invention can accurately, quickly, and efficiently detect leaks and their occurrence, and can monitor the underground natural gas gathering and transmission pipeline in real time during operation to effectively prevent explosions caused by leaks in the underground natural gas gathering and transmission pipeline.

[0018] This invention establishes leakage models for buried natural gas pipelines with different physical properties, acquiring data on leakage amount and time, leakage velocity and time, pressure and time before the leakage point, and pressure drop and time near the leakage point. It then utilizes a multinomial linear regression model from machine learning to process multi-source heterogeneous data fusion technology and generate a preliminary pipeline leakage prediction model. Error analysis is then performed on this preliminary model, resulting in a comprehensive prediction model for buried natural gas pipelines with different physical properties. This model provides detection technology support for the storage, transportation, and production of buried natural gas pipelines. It can efficiently and accurately pinpoint pipeline leakage points, exhibiting strong applicability and real-time performance, and is worthy of widespread adoption. Attached Figure Description

[0019] Figure 1 This is a block diagram of the present invention.

[0020] Figure 2 This is a flowchart illustrating the preliminary prediction model for leaks in buried natural gas gathering and transmission pipelines according to the present invention. Detailed Implementation

[0021] This invention discloses a method for predicting leakage in buried natural gas gathering and transmission pipelines, the specific method of which is as follows:

[0022] Using the finite element method combined with the service environment and working conditions of the buried natural gas pipeline, a leakage model with the same pipe diameter, different materials, burial depth, leakage hole, and temperature was established according to the principle of proportionality.

[0023] The proportion of the transport medium in the leakage model is set by the multi-component material parameter setting module, and different working conditions are simulated to establish a leakage simulation calculation model for buried natural gas gathering and transmission pipelines with different physical properties. Based on the leakage simulation calculation model of buried natural gas gathering and transmission pipelines with different physical properties, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are simulated and calculated to obtain the data of the pipeline leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point over time. A set of the above data is obtained at the required time interval to obtain the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties.

[0024] The simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties are cleaned, transformed and standardized by multi-source heterogeneous data fusion technology to obtain the consistent change relationship between leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point and time. The change characteristics of each physical quantity of buried natural gas gathering and transmission pipeline leakage are analyzed, and corresponding prediction conclusions of buried natural gas gathering and transmission pipeline leakage are given.

[0025] Based on the prediction conclusions corresponding to the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties, a training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model is generated, which includes leakage amount, leakage velocity, pressure before the leakage point, and pressure drop near the leakage point.

[0026] Using a multinomial linear regression model in machine learning, a training sample dataset for a leak prediction model of buried natural gas gathering and transmission pipelines was used to train and optimize the leak model, resulting in an optimized training sample dataset for the leak prediction model. The linear error of the training sample dataset was analyzed, and the training sample dataset for the pipeline leak prediction model regarding leakage amount, leakage velocity, pressure before the leak point, and pressure drop near the leak point was obtained, thus establishing a preliminary leak prediction model for buried natural gas gathering and transmission pipelines.

[0027] By conducting indoor leakage characteristic tests on buried natural gas gathering and transmission pipelines, the correctness and effectiveness of the preliminary leakage prediction model for buried natural gas gathering and transmission pipelines were verified. The characteristics of leakage results and the differences in prediction results under the same test conditions were analyzed. Error analysis, a mathematical analysis method, was used to analyze the leakage prediction error of buried natural gas gathering and transmission pipelines, and a test sample of the pipeline leakage prediction model was obtained. The preliminary leakage prediction model for buried natural gas gathering and transmission pipelines was then optimized to obtain a new leakage prediction model. When leakage prediction is required, the operating parameters of the buried natural gas pipeline, such as pipeline operating pressure, gas flow velocity, and pressure drop, are input into the prediction model. The model is then compared and analyzed with the training sample dataset of the pipeline leakage prediction model to determine whether a leakage has occurred in the pipeline.

[0028] In practical applications, the buried natural gas gathering and transmission pipeline leakage prediction model will be developed into a corresponding program, and an application program and a human-computer interaction interface for pipeline leakage prediction will be established in the computer system.

[0029] When establishing a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the Fluent fluid field module in ANSYS Workbench was selected as the modeling and simulation environment. The Mixture Material multi-component material parameter setting module was used to set the proportions of CH4, H2S, H2O, CO2, O2, C2H6, H2, and N2 components in the natural gas pipeline. According to the actual size and design specifications of the buried natural gas gathering and transmission pipeline, a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties was established.

[0030] The data on pipeline leakage rate, leakage velocity, pressure upstream of the leak point, and pressure drop near the leak point over time are collected in the following manner:

[0031] Based on the establishment of a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the inlet pressure and velocity parameters of multi-component gas are set at the front end of the natural gas pipeline, and the outlet pressure and velocity parameters of multi-component gas are set at the rear end of the natural gas pipeline. The porosity, viscous resistance coefficient and inertial resistance coefficient of the soil outside the buried natural gas gathering and transmission pipeline are set. Through simulation calculation, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are obtained from the simulation calculation model of leakage of buried natural gas gathering and transmission pipeline over time.

[0032] The training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model includes training sample datasets on the changes in leakage amount and time, the changes in leakage velocity and time, the changes in pressure before the leakage point and time, and the changes in pressure drop near the leakage point and time.

[0033] A leakage prediction model for buried natural gas gathering and transmission pipelines is provided, which is obtained using one of the aforementioned leakage prediction methods for buried natural gas gathering and transmission pipelines.

Claims

1. A method for predicting leaks in buried natural gas gathering and transmission pipelines, characterized in that, The specific method is as follows: Using the finite element method combined with the service environment and working conditions of the buried natural gas pipeline, a leakage model with the same pipe diameter, different materials, burial depth, leakage hole, and temperature was established according to the principle of proportionality. The proportion of the transport medium in the leakage model is set by the multi-component material parameter setting module, and different working conditions are simulated to establish a leakage simulation calculation model for buried natural gas gathering and transmission pipelines with different physical properties. Based on the leakage simulation calculation model of buried natural gas gathering and transmission pipelines with different physical properties, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are simulated and calculated to obtain the data of the pipeline leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point over time. A set of the above data is obtained at the required time interval to obtain the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties. The simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties are cleaned, transformed and standardized by multi-source heterogeneous data fusion technology to obtain the consistent change relationship between leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point and time. The change characteristics of each physical quantity of buried natural gas gathering and transmission pipeline leakage are analyzed, and corresponding prediction conclusions of buried natural gas gathering and transmission pipeline leakage are given. Based on the prediction conclusions corresponding to the simulation calculation results of buried natural gas gathering and transmission pipelines with different physical properties, a training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model is generated, which includes leakage amount, leakage velocity, pressure before the leakage point, and pressure drop near the leakage point. Using a multinomial linear regression model in machine learning, a training sample dataset for a leak prediction model of buried natural gas gathering and transmission pipelines was used to train and optimize the leak model, resulting in an optimized training sample dataset for the leak prediction model. The linear error of the training sample dataset was analyzed, and the training sample dataset for the pipeline leak prediction model regarding leakage amount, leakage velocity, pressure before the leak point, and pressure drop near the leak point was obtained, thus establishing a preliminary leak prediction model for buried natural gas gathering and transmission pipelines. By conducting indoor leakage characteristic tests on buried natural gas gathering and transmission pipelines, the correctness and effectiveness of the preliminary leakage prediction model for buried natural gas gathering and transmission pipelines were verified. The characteristics of leakage results and the differences in prediction results under the same test conditions were analyzed. Error analysis, a mathematical analysis method, was used to analyze the leakage prediction error of buried natural gas gathering and transmission pipelines, and the preliminary leakage prediction model was optimized. The resulting buried natural gas gathering and transmission pipeline leakage prediction model is then obtained. When leakage prediction is needed, the operating parameters of the buried natural gas pipeline are input into the prediction model. The model is then compared and analyzed with the training sample dataset of the pipeline leakage prediction model to determine whether a leak has occurred in the pipeline.

2. The method for predicting leakage in buried natural gas gathering and transmission pipelines as described in claim 1, characterized in that, When establishing simulation calculation models for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the Fluent fluid field module in ANSYS Workbench was selected as the modeling and simulation environment. The Mixture Material multi-component material parameter setting module was used to set the proportions of CH4, H2S, H2O, CO2, O2, C2H6, H2, and N2 components in the natural gas pipeline. According to the actual size and design specifications of the buried natural gas gathering and transmission pipeline, simulation calculation models for leakage of buried natural gas gathering and transmission pipelines with different physical properties were established.

3. The method for predicting leakage in buried natural gas gathering and transmission pipelines as described in claim 2, characterized in that, The data on pipeline leakage rate, leakage velocity, pressure upstream of the leak point, and pressure drop near the leak point over time are collected in the following manner: Based on the establishment of a simulation calculation model for leakage of buried natural gas gathering and transmission pipelines with different physical properties, the inlet pressure and velocity parameters of multi-component gas are set at the front end of the natural gas pipeline, and the outlet pressure and velocity parameters of multi-component gas are set at the rear end of the natural gas pipeline. The porosity, viscous resistance coefficient and inertial resistance coefficient of the soil outside the buried natural gas gathering and transmission pipeline are set. Through simulation calculation, the leakage amount, leakage velocity, pressure before the leakage point and pressure drop near the leakage point are obtained from the simulation calculation model of leakage of buried natural gas gathering and transmission pipeline over time.

4. The method for predicting leakage in buried natural gas gathering and transmission pipelines as described in claim 3, characterized in that, The training sample dataset for the buried natural gas gathering and transmission pipeline leakage prediction model includes training sample datasets on the changes in leakage amount and time, training sample datasets on the changes in leakage velocity and time, training sample datasets on the changes in pressure before the leakage point and time, and training sample datasets on the changes in pressure drop near the leakage point and time.

5. A leakage prediction model for buried natural gas gathering and transmission pipelines, characterized in that, The model was obtained using a method for predicting leaks in buried natural gas gathering and transmission pipelines as described in any one of claims 1-4.

Images