Shaft flow characteristic identification method based on combination of optical fiber monitoring and numerical simulation

Abstract

The invention discloses a shaft flow characteristic identification method based on combination of optical fiber monitoring and numerical simulation, and belongs to the technical field of multiphase flow correlation, and the method comprises the following steps: firstly, building a DAS monitoring shaft multiphase flow pattern experiment platform; secondly, establishing a three-dimensional model by utilizing SpaceClaim according to an experimental wellbore model, importing the model into a mesh to perform mesh division, importing a mesh model into Fluent to set simulation conditions and perform simulation, fitting a simulation result with an experimental result, and continuously correcting the simulation model; and finally, according to the corrected simulation result and the experimental test result, key factors influencing the fluid flow pattern are summarized, and powerful technical support is provided for analysis of the fluid flow pattern in the shaft through a machine learning algorithm. The method is used for solving the problems that in the prior art, fluid flow pattern recognition in a shaft is inaccurate, incomplete, not timely and the like.

Description

Technical Field

[0001] The present invention relates to the technical field related to multiphase flow, and in particular to a method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation. Background Art

[0002] With the continuous development of fiber-optic monitoring technology, the application of distributed fiber optics is increasing across major oil and gas fields. Fiber-optic monitoring and sensing technology offers advantages such as high sensitivity, high resolution, and strong anti-interference capabilities. It also utilizes the Doppler effect flow velocity measurement principle to achieve long-distance distributed flow velocity sensing in various spatial environments. However, in the oil and gas production process, traditional monitoring and identification methods have limitations and are unable to provide comprehensive, dynamic, and real-time monitoring of fluid flow patterns within the wellbore. Summary of the Invention

[0003] The purpose of the present invention is to provide a wellbore flow characteristic identification method based on the combination of optical fiber monitoring and numerical simulation. Based on distributed optical fiber acoustic sensing (DAS) monitoring technology and Fluent numerical simulation technology, a comprehensive study of the fluid flow characteristics in the wellbore is carried out, the experimental results and numerical simulation results are compared and analyzed, and the key factors affecting the fluid flow pattern are summarized. This not only improves the accuracy and stability of the monitoring data, but also enhances the safety and efficiency of oil production operations, which is of great significance for realizing the intelligence of oil and gas fields.

[0004] To achieve the above objectives, the present invention provides a method for identifying wellbore flow characteristics based on a combination of optical fiber monitoring and numerical simulation, comprising the following steps:

[0005] S1. Build a DAS monitoring wellbore multiphase flow pattern experimental platform, set up wellbores with different inclination angles and different gas-liquid flow rates, and monitor the flow pattern of the fluid in the wellbore;

[0006] S2. Based on the experimental wellbore model, a 3D model is built using SpaceClaim. The model is imported into Mesh for meshing. The mesh model is imported into Fluent to set simulation conditions and perform simulation. The simulation results are fitted with the experimental results, and the simulation model is continuously revised.

[0007] S3. Based on the revised simulation results and experimental test results, summarize the key factors affecting the fluid flow pattern;

[0008] S4. Through machine learning algorithms, build an artificial neural network flow pattern recognition model to analyze experimental data and simulation data.

[0009] Preferably, the specific operation of S1 is:

[0010] S11. Build an experimental platform and wind the optical fiber along the pipe wall. Connect one end of the optical fiber to the DAS and leave the other end exposed to the air.

[0011] S12. Setting wellbores with different inclination angles and different gas and liquid flow rates, and visually observing and recording the specific flow patterns of the fluids in the wellbores;

[0012] S13. Record the changes in the waterfall diagram of DAS under different flow patterns.

[0013] Preferably, the specific operation of S2 is:

[0014] S21. According to the actual conditions in the laboratory, the basic parameters and dimensions of the wellbore are obtained, and wellbore models with different inclination angles are established using SpaceClaim.

[0015] S22, importing the wellbore model obtained in S21 into a mesh processor, dividing the wellbore model into meshes using a multi-zone meshing method, setting names of the inlet, outlet, and wall of the wellbore model, and outputting a mesh file;

[0016] S23, using the output mesh file for simulation analysis, setting the simulation mathematical model, physical parameters, boundary conditions, selecting an appropriate calculation method, starting numerical simulation calculation, and obtaining simulation results;

[0017] S24. After the simulation calculation is completed, use the Fluent post-processing tool to check the distribution of gas and liquid in the wellbore, observe and record the flow pattern data, and obtain the velocity vector diagram and phase distribution diagram; import the wellbore model with different inclination angles, change the flow rate at the inlet, keep other conditions unchanged, and then simulate again, and record the velocity vector diagram and phase distribution diagram;

[0018] S25. Compare the data recorded in the simulation with the experimental data. If the error between the data is within the allowable range, it is considered that the various conditions and grid division methods used in the simulation are reliable. If the error between the data exceeds the allowable range, repeat steps S21-S24 to correct the grid division, mathematical model and boundary condition setting until the error is within the error range.

[0019] Preferably, the specific operation of S3 is:

[0020] S31. Observe the changes in fluid flow patterns in the wellbore with the same gas-liquid two-phase flow rate at the inlet and different inclination angles;

[0021] S32. Keeping the wellbore inclination angle the same, change the gas-liquid two-phase flow rate at the inlet and observe the changes in the fluid flow pattern in the wellbore.

[0022] Preferably, the specific operation of S4 is:

[0023] S41. Select experimental data from a gas-liquid two-phase flow pattern measurement experiment, design a flow pattern recognition model using a BPN artificial neural network based on the relationship between the gas-liquid two-phase superficial velocity, pipeline inclination, and flow pattern, and train and test the model.

[0024] S42. Based on the artificial neural network flow pattern recognition model, the gas phase apparent velocity, liquid phase apparent velocity, and pipeline inclination are used as input variables, and the flow pattern category is used as the output variable to obtain the overall recognition accuracy.

[0025] Therefore, the present invention adopts the above-mentioned wellbore flow feature identification method based on the combination of optical fiber monitoring and numerical simulation. First, a DAS monitoring wellbore multiphase flow pattern experimental platform is built, wellbores with different inclination angles and different gas-liquid flow rates are set, and the flow state of the fluid in the wellbore is monitored; secondly, according to the experimental wellbore model, a three-dimensional model is established using SpaceClaim, the model is imported into mesh for grid division, the mesh model is imported into Fluent to set simulation conditions and simulate, the simulation results are fitted with the experimental results, and the simulation model is continuously corrected to match the experimental results as much as possible; finally, according to the corrected simulation results and experimental test results, the key factors affecting the fluid flow pattern are summarized, and the experimental data and simulation data are analyzed based on the artificial neural network flow pattern recognition model through machine learning algorithm, providing strong technical support for the analysis of fluid flow pattern in the wellbore. The present invention provides a new monitoring method for analyzing the identification of fluid flow pattern in the wellbore during drilling, which is used to solve the problems of inaccurate, incomplete and untimely identification of fluid flow pattern in the wellbore in the prior art;

[0026] Using DAS as a fluid flow pattern monitoring tool, and taking advantage of its high precision and real-time monitoring capabilities, the multiphase flow pattern in the wellbore is monitored and analyzed. Through the DAS system, the acoustic signal data of the fluid in the wellbore can be continuously obtained to capture the characteristics and changes of the fluid flow. Compared with traditional fluid flow pattern identification technology, it has higher spatial resolution, longer monitoring distance, and real-time monitoring capabilities, which is more efficient, accurate and safe. Combined with the Fluent simulation model and machine learning algorithm, the present invention can effectively identify and analyze the flow pattern of the fluid in the wellbore, optimize the monitoring effect, and provide precise technical support for fluid dynamics research.

[0027] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of an embodiment of a method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation according to the present invention;

[0029] Figure 2This is a diagram of an experimental platform constructed according to an embodiment of a method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation;

[0030] Figure 3 This is a first schematic diagram of wellbore grid division according to an embodiment of a wellbore flow feature identification method based on the combination of optical fiber monitoring and numerical simulation of the present invention;

[0031] Figure 4 This is a second schematic diagram of wellbore grid division according to an embodiment of a wellbore flow feature identification method based on the combination of optical fiber monitoring and numerical simulation of the present invention;

[0032] Figure 5 This is a simulation flow pattern diagram of an embodiment of a wellbore flow characteristic identification method based on the combination of optical fiber monitoring and numerical simulation of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0034] Unless otherwise defined, technical or scientific terms used in the present invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0035] Example 1

[0036] like Figure 1 As shown, the present invention provides a method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation, comprising the following steps:

[0037] S1. Build Figure 2 The DAS monitoring wellbore multiphase flow pattern experimental platform shown in the figure sets wellbores with different inclination angles and different gas-liquid flow rates to monitor the flow pattern of the fluid in the wellbore. Its specific operations are as follows:

[0038] S11. Build an experimental platform and wind the optical fiber along the pipe wall. Connect one end of the optical fiber to the DAS and leave the other end exposed to the air.

[0039] S12. Setting wellbores with different inclination angles and different gas and liquid flow rates, and visually observing and recording the specific flow patterns of the fluids in the wellbores;

[0040] S13. Record the changes in the waterfall diagram of DAS under different flow patterns.

[0041] S2. Based on the experimental wellbore model, use SpaceClaim to build a three-dimensional model and import the model into mesh for mesh division, such as Figure 3 and Figure 4As shown, the grid model is imported into Fluent to set simulation conditions and perform simulation. The simulation results are fitted with the experimental results, and the simulation model is continuously revised. The specific operations are as follows:

[0042] S21. According to the actual conditions in the laboratory, the basic parameters and dimensions of the wellbore are obtained, and wellbore models with different inclination angles are established using SpaceClaim.

[0043] S22, importing the wellbore model obtained in S21 into a mesh processor, dividing the wellbore model into meshes using a multi-zone meshing method, setting names of the inlet, outlet, and wall of the wellbore model, and outputting a mesh file;

[0044] S23, using the output mesh file for simulation analysis, setting the simulation mathematical model, physical parameters, boundary conditions, selecting an appropriate calculation method, starting numerical simulation calculation, and obtaining simulation results;

[0045] S24. After the simulation calculation is completed, use the Fluent post-processing tool to check the distribution of gas and liquid in the wellbore, observe and record the flow pattern data, and obtain the velocity vector diagram and phase distribution diagram; import the wellbore model with different inclination angles, change the flow rate at the inlet, keep other conditions unchanged, and then simulate again, and record the velocity vector diagram and phase distribution diagram;

[0046] S25. Compare the data recorded in the simulation with the experimental data. If the error between the data is within the allowable range, it is considered that the various conditions and grid division methods used in the simulation are reliable. If the error between the data exceeds the allowable range, repeat steps S21-S24 to correct the grid division, mathematical model and boundary condition setting until the error is within the error range.

[0047] S3. Based on the revised simulation results and experimental test results, the key factors affecting the fluid flow pattern are summarized. The simulation flow pattern diagram is as follows: Figure 5 As shown, the specific operations are:

[0048] S31. Observe the changes in fluid flow patterns in the wellbore with the same gas-liquid two-phase flow rate at the inlet and different inclination angles;

[0049] S32. Keeping the wellbore inclination angle the same, change the gas-liquid two-phase flow rate at the inlet and observe the changes in the fluid flow pattern in the wellbore.

[0050] S4. Using machine learning algorithms, we build an artificial neural network flow pattern recognition model to analyze experimental and simulation data. The specific operations are as follows:

[0051] S41. Select experimental data from a gas-liquid two-phase flow pattern measurement experiment, design a flow pattern recognition model using a BPN artificial neural network based on the relationship between the gas-liquid two-phase superficial velocity, pipeline inclination, and flow pattern, and train and test the model.

[0052] S42. Based on the artificial neural network flow pattern recognition model, the gas phase apparent velocity, liquid phase apparent velocity, and pipeline inclination are used as input variables, and the flow pattern category is used as the output variable to obtain the overall recognition accuracy.

[0053] Therefore, the present invention adopts the above-mentioned wellbore flow characteristic identification method based on the combination of optical fiber monitoring and numerical simulation, which provides a new monitoring method for analyzing the flow pattern of fluid in the wellbore during drilling, and is used to solve the problems of inaccurate, incomplete and untimely identification of fluid flow pattern in the wellbore in the existing technology.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation, characterized in that: The following steps are involved: S1. Build a DAS monitoring wellbore multiphase flow pattern experimental platform, set up wellbores with different inclination angles and different gas-liquid flow rates, and monitor the flow pattern of fluids in the wellbore; S2. Based on the experimental wellbore model, use SpaceClaim to build a three-dimensional model, import the model into mesh for meshing, import the mesh model into Fluent to set simulation conditions and perform simulation, fit the simulation results with the experimental results, and continuously modify the simulation model; S3. Based on the revised simulation results and experimental test results, the key factors affecting the fluid flow pattern are summarized; S4. Through machine learning algorithms, build an artificial neural network flow pattern recognition model to analyze experimental data and simulation data.

2. A method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation according to claim 1, characterized in that: The specific operation of S1 is: S11. Build an experimental platform, wind the optical fiber along the pipe wall, connect one end of the optical fiber to the DAS, and expose the other end to the air; S12, setting wellbores with different inclination angles and different gas-liquid flow rates, and visually observing and recording the specific conditions of the fluid flow patterns in the wellbores; S13. Record the changes in the waterfall diagram of DAS under different flow patterns.

3. The method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation according to claim 1 is characterized in that: The specific operation of S2 is: S21. According to the actual situation in the laboratory, the basic parameters and dimensions of the wellbore are obtained, and wellbore models with different inclination angles are established using SpaceClaim; S22, importing the wellbore model obtained in S21 into a mesh processor, dividing the wellbore model into meshes by a multi-zone meshing method, setting the names of the inlet, outlet and wall of the wellbore model, and outputting a mesh file; S23, using the output mesh file for simulation analysis, setting the simulation mathematical model, physical parameters, boundary conditions, selecting a suitable calculation method, starting numerical simulation calculation, and obtaining simulation results; S24. After the simulation calculation is completed, use the Fluent post-processing tool to check the distribution of gas and liquid in the wellbore, observe and record the flow pattern data, and obtain the velocity vector diagram and phase distribution diagram; import the wellbore model with different inclination angles, change the flow rate at the inlet, keep other conditions unchanged and then simulate, and also record the velocity vector diagram and phase distribution diagram; S25. Compare the simulation recorded data with the experimental data. If the error between the data is within the allowable range, it is considered that the various conditions and grid division methods used in the simulation are reliable. If the error between the data exceeds the allowable range, repeat steps S21-S24 to correct the grid division, mathematical model and boundary condition setting until the error is within the error range.

4. The method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation according to claim 1, characterized in that: The specific operation of S3 is: S31, observe the change of fluid flow pattern in the wellbore with the same gas-liquid two-phase flow rate and different inclination angles at the inlet; S32. Keeping the wellbore inclination angle the same, change the gas-liquid two-phase flow rate at the inlet and observe the changes in the fluid flow pattern in the wellbore.

5. The method for identifying wellbore flow characteristics based on the combination of optical fiber monitoring and numerical simulation according to claim 1, characterized in that: The specific operation of S4 is: S41. Select experimental data of gas-liquid two-phase flow pattern measurement experiment, design a flow pattern recognition model using BPN artificial neural network according to the relationship between gas-liquid two-phase superficial velocity, pipeline inclination and flow pattern, and train and test the model; S42. Based on the artificial neural network flow pattern recognition model, the gas phase apparent velocity, liquid phase apparent velocity, and pipeline inclination are used as input variables, and the flow pattern category is used as the output variable to obtain the overall recognition accuracy.

Images