Optical fiber integrated with numerical simulator for pipeline leakage quantification
By combining optical fiber sensing and numerical simulator, the problem of difficult to accurately detect and quantify pipeline leakage in the prior art is solved, and the accurate identification and quantification of leakage location, size and severity is achieved, and detection accuracy and rapid response are improved. ability.
Patent Information
- Application Number
- CN202310655625.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-05
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art is difficult to accurately detect and quantify the location, size and severity of oil and gas pipeline leakage. Fiber sensors can detect the location of leakage but cannot explain the size and severity, while numerical simulators are difficult to predict the location of leakage separately.
Combining fiber sensing and numerical simulators, detect abnormalities by installing fiber cables outside the pipeline, numerical simulators to evaluate leakage location, size, and traffic, generate downgrade machine learning models to quickly identify leakages, and verify and predict futures with prediction results change.
Accurate identification and quantification of pipeline leakage is achieved, detection accuracy is improved, false alarms are reduced, and rapid and quantitative insights are provided for easy decision-making.
Smart Images

Figure CN119084842A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates generally to quantification of pipeline leakage and, more particularly, to integrated fiber optics and numerical simulators for pipeline leakage quantification. Background Art
[0002] Although oil and gas pipeline networks are considered the most economical and safest means of transporting crude oil efficiently and reliably, the number of serious incidents caused by pipeline failures has continued to increase in recent decades. Such failures can be intentional (vandalism) or unintentional (equipment / material failure and corrosion) damage, leading to catastrophic and irreversible effects such as ecological disasters, economic losses and extreme environmental pollution. This is especially true when pipeline leaks are not detected in time. Summary of the invention
[0003] In some configurations, a method includes detecting a leakage event using fiber optic sensing; and evaluating the leakage event using a numerical simulator.
[0004] In some configurations, a method includes preparing a base simulation case in which no leak exists; detecting the leak using fiber optic sensing; inputting the leak location into a numerical simulator; and performing a parametric study of the leak location to identify a size and flow rate of the leak.
[0005] In some configurations, a method includes preparing a base simulation case in which no leak exists; generating a large database by running parameter study simulations; building a reduced-order machine learning model from the database; detecting the leak using fiber optic sensing; and inputting the leak location into the reduced-order machine learning model.
[0006] The method may also include using the forecast results to run a forward model to further validate the explanation and predict upcoming changes. BRIEF DESCRIPTION OF THE DRAWINGS
[0007] Certain embodiments, features, aspects and advantages of the present disclosure will be described below with reference to the accompanying drawings, wherein the same reference numerals represent the same elements. It should be understood that the accompanying drawings illustrate various implementations described herein and are not intended to limit the scope of the various technologies described herein.
[0008] Figure 1 An example pipeline monitoring system is schematically illustrated.
[0009] Figure 2 An example workflow integrating optical fiber and numerical simulation is shown.
[0010] Figure 3 A demonstration simulation case of pipeline leakage monitoring is schematically shown.
[0011] Figure 4 A graph of inlet pressure versus time is shown for leak diameters ranging from 2% to 20% of the pipe diameter.
[0012] Figure 5 A graph showing the outlet flow rate versus time for leak diameters ranging from 1% to 20% of the pipe diameter is shown.
[0013] Figure 6 A graph showing the variation of leakage rate at different leakage diameters is shown.
[0014] Figure 7 A graph showing the pressure variation along the pipeline for each leak size under steady state conditions is shown.
[0015] Figure 8 An example workflow for real-time integration of optical fiber and numerical simulation by generating a reduced-order model is shown. DETAILED DESCRIPTION
[0016] In the following description, many details are set forth to provide an understanding of some embodiments of the present disclosure. It should be understood that the following disclosure provides many different embodiments or examples for realizing the different features of various embodiments. Specific examples of components and arrangements are described below to simplify the disclosure. Of course, these are merely examples and are not intended to be limiting. However, it will be appreciated by those of ordinary skill in the art that the system and / or method can be implemented without these details, and that many variations or modifications to the described embodiments are possible. This description should not be construed as restrictive, but only for the purpose of describing the general principles of the embodiments. The scope of the described embodiments should be determined with reference to the claims issued.
[0017] As used herein, the terms "connecting," "connecting," "connected," "connected with," and "connected with" are used to mean "directly connected with" or "connected with via one or more elements"; the term "group" is used to mean "one element" or "more than one element." In addition, the terms "connecting," "connecting," "connected," "coupled together," and "coupled with" are used to mean "directly coupled together" or "coupled together via one or more elements." The terms "upper" and "lower"; "above"; "below"; "top" and "bottom"; and other similar terms indicating positions relative to a given point or element are used to more clearly describe some elements. Generally, these terms relate to a reference point on the ground from which the drilling operation is initiated as the top point and the total depth is the lowest point, where the well (e.g., wellbore, borehole) is vertical, horizontal, or inclined relative to the ground.
[0018] A great deal of research effort has been devoted to the detection and location of leaks in oil and gas pipelines using various methods. Two specific methods are fiber optic sensing and dynamic numerical simulation. Remote fiber optic sensing technology is often used to mitigate a leak event or identify the leak location so operators can quickly take data-driven actions to minimize the severity. This technology requires the installation of fiber optic sensors on the outside of the pipeline. When a pipeline leak occurs and hydrocarbon fluids affect the coated cable measurements, acoustic and temperature anomalies in the coated cable sensing can be detected. This method is generally applicable to above ground gathering networks, buried intercontinental oil and gas pipelines, and is applicable to all fluid types.
[0019] Dynamic numerical simulation is a computational method that primarily requires sensor measurements of pressure, flow and temperature at several locations along the pipeline. Based on physical principles, mathematical models are formulated to represent the operation of the pipeline system. Computationally intensive simulations are performed to detect leaks using the differences between measured data and simulated values, based on conservation equations and state equations for the fluid. This is sometimes referred to as real-time transient modeling (RTTM).
[0020] When a fiber optic solution effectively detects a leak event, it reports a relatively accurate leak location, but cannot physically explain the size of the leak and the severity of the event. On the other hand, numerical simulators are often difficult to predict leak location alone, but they do provide the ability to understand the event through physical principles.
[0021] The present disclosure provides a system and method for integrating optical fiber and numerical simulators to identify and quantify pipeline leaks. Optical fiber cables are installed along the pipeline to track fluid flow and noise, identify anomalies, and perform event classification, including scenarios that identify relatively accurate pipeline leak locations. The detected leak locations are then used in a numerical simulator that also considers pressure, temperature, and flow measurements at discrete locations along the pipeline to further assess, for example, leak size, flow and volume loss, and therefore assess the severity of the event.
[0022] Figure 1 An example monitoring system according to the present disclosure is schematically shown. As shown, the system includes a fiber optic cable installed along the pipeline (e.g., along the outside of the pipeline), at least two sets of devices (e.g., sensors) for measuring pressure, temperature and / or flow at the pipeline inlet and outlet, and a representative pipeline simulation condition without leakage. The inlet flow, inlet temperature and outlet pressure are controlled parameters, represented by solid boxes. The dashed boxes represent characteristics that vary based on fluid dynamic conditions, such as inlet pressure, outlet temperature and outlet flow. Figure 2An example workflow according to the present disclosure, integrated fiber optic and numerical simulator technology is shown. As shown, the workflow may include one or more of the following steps: preparing a basic simulation case when no leak exists; using fiber optics for event classification and / or leak location; inputting the leak location into the numerical simulator; and performing a parameter study on a known leak location to determine the size and flow of the leak.
[0023] 1. Basic simulation when there is no leak
[0024] Before leak detection begins, prepare a basic simulation scenario. The basic scenario covers the basic parameters of the pipeline setup, such as fluid composition and PVT characteristics, pipe flow path and geometry (diameter, length, etc.), and / or control of inlet and outlet pressure, temperature and flow. Figure 3 The schematic diagram shows a demonstration simulation of a simple pipeline structure with pure CO2 content. Some parameters of the example setup are shown below:
[0025] Pipeline: Length = 40km, Elevation = 0m, Diameter = 0.2m, Roughness = 2.8e-5m, Number of pipe segments = 20 (for plotting results), Pipe wall material parameters such as capacity, conductivity, density, etc.
[0026] ·Ingredients: Single ingredient CO2.
[0027] Inlet control: mass source, temperature = 10 degrees Celsius, mass flow rate = 15 kg / sec.
[0028] Outlet control: pressure driven, pressure = 20.68 bar.
[0029] Leakage source: valve type, section (sevtion) ∈ [1, 20], leakage back pressure = 0 bar, discharge coefficient = 0.84, equivalent diameter of the leakage area ∈ [0.002, 0.04] m, that is, between 1% and 20% of the pipe diameter.
[0030] A basic simulation scenario was constructed before deploying the solution described in this article. The leak source setting was initially a dummy setting with a leak diameter of zero.
[0031] 2. Event classification and leak location using optical fiber
[0032] Real-time acoustic and temperature data are processed through complex analytical workflows and machine learning techniques. The output of this step includes one or more of the following: 1) the type of event detected, which can be digging, drilling, excavator, fiber break, pigging, etc.; 2) the severity of the event, which can be a rough classification of low, medium and high; this classification has no direct physical meaning, but is determined by a threshold of abnormal level; and 3) the location of the event, which can be accurate to a few meters.
[0033] 3. Input the leak location into the numerical simulator
[0034] When a leak event is detected by fiber optic technology, it reports a relatively accurate leak location. However, fiber optic technology cannot physically explain the size of the leak and the severity of the event. On the other hand, numerical simulators often have difficulty predicting the leak location alone. However, numerical simulators do provide the ability to understand the event through physics.
[0035] In the example case, the leak back pressure and discharge coefficient parameters are fixed. Without knowing the leak location, one can try a parametric study that varies the diameter and section parameters as shown below:
[0036]
[0037] This will create a parameter study of 20x20=400 cases. A more serious problem with this approach, besides performance, is that it is challenging to identify the most likely case out of the 400 cases, which may reduce the accuracy of the inversion problem.
[0038] By inputting the leak location into the numerical simulator, simply run 20 cases as shown below.
[0039]
[0040] In many cases, the behavior of pressure and flow rate for different leak diameters can have a monotonic behavior, which significantly reduces the uncertainty of the inversion.
[0041] In more complex flow simulations, the leakage backpressure (ambient pressure) and discharge coefficient parameters may need to be included in the parameter study, in which case it will be a 4D matrix as shown below. In this example, the parameter study count is intentionally reduced, but 6^4 = 1296 cases are still required.
[0042]
[0043] 4. Parametric studies of size and flow rate to identify leaks
[0044] After running the model with the parameter matrix, a series of simulated physical properties can be compared with actual measurements to identify the most likely scenario. In this example, the inlet flow rate and outlet pressure are controlled, so simulation results for inlet pressure and outlet flow rate at different leak diameters can be generated. Figure 4 The inlet pressure is shown as a function of time for leak diameters ranging from 2% to 20% of the pipe diameter. It can be observed that when there is no leak, the inlet pressure is approximately 45 bar. As the leak becomes larger, the pressure gradually drops to 33.5 bar. While in this simple case the correlation between inlet pressure and leak diameter is monotonic, it can be more complex for other simulation cases. Figure 5 The outlet mass flow rate is shown as a function of time for leak diameters ranging from 1% to 20% of the pipe diameter.
[0045] The most plausible leak size can be determined by performing a similarity analysis (pattern recognition) between the simulated pressure and flow curves and the actual measured values. This similarity analysis can be done using existing statistical methods or computer vision methods. The corresponding leakage flow can be trivially determined by mass balance, e.g. Figure 6 In some cases, there may be additional measurements between the inlet and outlet that can also be used for comparison with the simulation results. For example, Figure 7 The pressure profile (pressure variation along the pipeline) is shown for each leak size at steady state. Therefore, any measurement along the pipeline will help determine the most plausible leak size.
[0046] 5. Reduced-Order Models for Fast Estimation
[0047] In some cases, pipeline leak detection requires fast response time, and a large number of parameter studies using numerical simulators may not be sufficient to quickly provide the information needed for decision making. The system and method of the present disclosure can be further enhanced by pre-generating a large database and only running the reduced-order model when a leak event is detected. The example specific workflow is as follows:
[0048] 1) Establish a baseline condition of no leaks for a specific pipeline facility.
[0049] 2) Generate a database by running multiple (e.g., thousands) parametric study simulations with different leak diameters, leak locations, leak back pressures, and discharge coefficients; the outputs of the simulations are inlet pressure, outlet flow and temperature, and leak rate. These records are recorded in a database.
[0050] -The inputs for each simulation are also shown in the table below:
[0051]
[0052] - Output: Inlet pressure, outlet flow and temperature, and leakage flow, all as a function of time.
[0053] - Please note that the above list assumes that the inlet flow and outlet pressure are controlled relatively stably over time. However, when operating conditions deviate from the control parameters, the simulation database should be rebuilt.
[0054] 3) Build a reduced-order machine learning model from the database. Note that the ML model has opposite inputs and outputs because we are now trying to predict leakage information.
[0055] - Reduced order model inputs: inlet pressure, outlet flow and temperature over time, plus leak location / section.
[0056] -Output: leak diameter, leak flow, back pressure and flow coefficient.
[0057] 4) When a leak event is detected, the leak location / segment identified by the optical fiber, together with the measured inlet pressure, outlet flow rate and temperature changes over time, are input into the reduced-order model, which is used to predict the most reasonable leak diameter and leak flow rate.
[0058] 5) Optionally, use the predictions to run additional forward models to further validate the interpretation and predict upcoming changes.
[0059] Figure 8 A flow chart summarizing the workflow using a reduced order model is shown.
[0060] This workflow is provided as an example to illustrate the concept rather than to provide a definitive list of inputs and outputs. For example, the leakage backpressure and flow coefficient may be fixed values to reduce computational cost, or the flow restriction may be changed from "inlet flow and outlet pressure" to "inlet pressure and outlet flow", in which case the list of related properties needs to be adjusted accordingly.
[0061] The systems and methods of the present disclosure advantageously combine optical fiber and numerical simulators to provide a leak detection solution with improved accuracy and reduced false alarms, thereby providing operators with rapid and quantitative insights for decision making.
[0062] Language of degree as used herein, such as the terms "approximately," "about," "substantially," and "substantially" as used herein, refers to a value, amount, or feature that is close to a stated value, amount, or feature that still performs a desired function or achieves a desired result. For example, the terms "approximately," "about," "substantially," and "substantially" may refer to an amount that is less than 10%, less than 5%, less than 1%, less than 0.1%, and / or less than 0.01% of a stated amount. As another example, in certain embodiments, the terms "substantially parallel" and "substantially parallel" or "substantially perpendicular" and "substantially perpendicular" refer to a value, amount, or feature that deviates from being completely parallel or perpendicular by less than or equal to 15 degrees, 10 degrees, 5 degrees, 3 degrees, 1 degree, or 0.1 degrees, respectively.
[0063] Although some embodiments of the present disclosure have been described in detail above, it will be readily appreciated by those skilled in the art that many modifications are possible without departing substantially from the teachings of the present disclosure. Therefore, such modifications are intended to be included within the scope of the present disclosure as defined in the claims. It is also conceivable that various combinations or sub-combinations may be made to the specific features and aspects of the described embodiments and still fall within the scope of the present disclosure. It should be understood that the various features and aspects of the disclosed embodiments may be combined or replaced with one another to form different modes of embodiments of the present disclosure. Therefore, it is intended that the scope of the present disclosure herein should not be limited by the specific embodiments described above.
Claims
1. A method comprising: Use fiber optic sensing to detect leak events; and A numerical simulator is used to evaluate leak events.
2. A method comprising: preparing a basic simulation case in which no leakage exists; Use fiber optic sensing to detect leaks; inputting the leak location into a numerical simulator; and Perform a parametric study of the leak location to identify the size and flow rate of the leak.
3. A method comprising: preparing a basic simulation case in which no leakage exists; Generate large databases by running parametric study simulations; constructing a reduced-order machine learning model from the database; Use fiber optic sensing to detect leaks; and The leak locations are fed into a reduced-order machine learning model.
4. The method of claim 3 further comprises running a forward model using the prediction results to further validate the explanation and predict upcoming changes.
Citation Information
Cited By
Fiber optics integrated with numerical simulators for pipeline leak quantification
US20240402032A1
