A tracking and verification method for gas reservoir-wellbore-surface integrated simulation model

Abstract

The present invention discloses a gas reservoir-wellbore-ground integrated simulation model tracking and verification method, and relates to the technical field of oil and gas field development. The present invention uses the integrated simulation model prediction value to track and compare with the real-time information obtained by the ground system sensor, and issues an early warning for the tracking index that exceeds the early warning threshold; when an alarm occurs, the fault location program is started to find the fault location, and the model is corrected for the fault location; the fault location program includes diagnosing gas reservoir and wellbore model faults with the bottom of the well as the solution node and diagnosing ground pipe network model faults with the wellhead as the solution node. Accurately locate the model fault location, so as to correct the model in a targeted manner, greatly improve the fitting efficiency of the integrated simulation model, and continuously ensure that the integrated simulation model is accurate and available.

Description

Technical Field

[0001] The invention relates to the technical field of oil and gas field development, and more specifically to a tracking and verification method for a gas reservoir-wellbore-ground integrated simulation model. Background Art

[0002] Gas field development is a complex and large production system, including three subsystems: formation seepage, wellbore lifting, and surface gathering and transportation. The integrated simulation model uses computer technology to establish gas reservoir, wellbore, and surface sub-models respectively, and then uses professional software to achieve coupling and parallel operations between models, truly reproducing the entire process of gas field development. To use an integrated model to guide production, it is necessary to ensure that the model is accurate in real time, but the integrated simulation model requires a huge and complex model parameter, and the model parameters change in real time with the development and production of the gas field. It is very difficult to ensure the accuracy of the integrated model in real time.

[0003] There is a corresponding industry standard for fitting gas reservoir models, "Technical Specifications for Application of Numerical Simulation of Oil and Gas Reservoirs", standard number SY / T 6744-2008. However, there is no confirmed high-precision fitting method for the integrated model of simulating the global production system. In 2019, there was a high-frequency fitting method for wellbore and IPR, "Correction Method, Device and Computer Reservoir Medium for Gas Well Inflow Dynamic Curve", patent application number 201910716909.X. This patent proposes a method for establishing a high-precision VLP multiphase pipe flow model and a real-time calibration method for the IPR inflow dynamic model. However, due to the coupled parallel computing characteristics of the integrated model, the sub-models affect each other. When the integrated model fails, it may be caused by any one or more sub-model parameter errors. The integrated model contains a huge number of parameters and changes in real time. Aimless adjustment and fitting will waste a lot of time. Summary of the invention

[0004] In order to overcome the defects and shortcomings in the above-mentioned prior art, the present invention provides a gas reservoir-wellbore-ground integrated simulation model tracking and verification method. The purpose of the present invention is to solve the problem that the integrated model fault cannot be accurately located and the adjustment and fitting efficiency is low in the above-mentioned prior art. The present invention first tracks and compares the predicted value of the integrated simulation model with the real-time information obtained by the ground system sensor to monitor the model fault in real time; establishes an integrated model fault positioning technology to accurately locate the model fault position, thereby correcting the model in a targeted manner, greatly improving the fitting efficiency of the integrated simulation model, and continuously ensuring the accuracy and availability of the integrated simulation model.

[0005] In order to solve the problems existing in the above-mentioned prior art, the present invention is achieved through the following technical solutions:

[0006] A gas reservoir-wellbore-ground integrated simulation model tracking and verification method comprises the following steps:

[0007] Step S1, using the predicted value of the integrated simulation model to track and compare with the real-time information obtained by the sensors in the actual ground system, and issuing an early warning for the tracking indicators that exceed the early warning threshold;

[0008] Step S2, when an alarm occurs in the production system simulated by the integrated simulation model, a fault location program is started to find the fault location and perform model correction on the fault location; the fault location program includes diagnosing gas reservoir model and / or wellbore model faults with the bottom of the well as the solution node, and diagnosing ground pipeline model faults with the wellhead as the solution node.

[0009] The step S1 is specifically as follows:

[0010] The integrated simulation model simulates the daily gas production of a fixed gas well, predicts the daily gas production of the gas well, the daily water production of the gas well, the wellhead pressure, the flow value of each metering point on the pipeline, and the pressure value of each metering point on the pipeline;

[0011] The actual ground system transmits the real-time information obtained by each sensor in the ground system to the integrated simulation model; the integrated simulation model tracks and compares the predicted values ​​of each indicator obtained based on the daily gas production simulation of the fixed gas well with the real-time information obtained by each sensor in the actual ground system. When the error between a certain tracking indicator and the real-time information exceeds the warning threshold of the tracking indicator, the production system simulated by the integrated simulation model alarms.

[0012] Furthermore, the method of diagnosing gas reservoir model and / or wellbore model faults by taking the well bottom as the solution node is specifically as follows:

[0013] Using the gas well pipe flow model, the daily gas production Q is calculated by simulating and calculating the wellhead pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. VLP ;

[0014] The daily gas production Q is calculated by using the dynamic characteristic model of gas well inflow and the formation pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. IPR ;

[0015] By analyzing the intersection of the inflow capacity curve and the outflow capacity curve, the gas production Q at the intersection point can be obtained. VLP / IPR ;

[0016] Comparison of daily gas production Q VLP 、Daily gas production Q IPR , gas production at intersection point Q VLP / IPR The actual daily gas production Q obtained from the real-time information obtained by the sensors of the actual ground system T , determine the fault location.

[0017] More specifically, if It is determined that the gas reservoir model has a fault;

[0018] like It is determined that the wellbore model has a fault; among them, Er Q is the maximum error allowed for the daily gas production of the integrated simulation model, which is also the warning threshold for the daily gas production.

[0019] Furthermore, the method of diagnosing the fault of the surface pipe network model with the wellhead as the solution node is specifically as follows:

[0020] The actual daily gas production, daily water production and joint station pressure at the wellhead are simulated using the surface pipeline network model to obtain the wellhead pressure P. C ; Compare the wellhead pressure P predicted by the surface pipeline network model C and the actual wellhead pressure P obtained from the real-time information obtained by various sensors of the actual surface system T , determine whether there is any fault in the ground pipe network model.

[0021] More specifically, if It is determined that there is a fault in the ground pipe network model, where Er P is the maximum error allowed for the wellhead pressure of the integrated simulation model, which is also the warning threshold of the wellhead pressure.

[0022] Furthermore, in step S1, the prediction value of the integrated simulation model is tracked and compared with the real-time information obtained by the sensors in the actual ground system, specifically including:

[0023] Step S101, determining tracking indicators;

[0024] Step S102: determining the warning threshold of each tracking indicator;

[0025] Step S103: Determine the tracking frequency.

[0026] The tracking indicators in step S101 include daily gas production of the gas well, daily water production of the gas well, wellhead pressure, flow values ​​of each metering point on the pipeline, and pressure values ​​of each metering point on the pipeline.

[0027] In step S102, there are two ways to determine the warning threshold of each tracking indicator. The first way is to determine the warning threshold by the fluctuation range of the tracking indicator during the historical production process of the gas well; the second way is to determine the warning threshold by using the stabilization time of the integrated simulation model tracking indicator; the maximum value of the above two ways is taken as the reasonable warning threshold.

[0028] The method of determining the warning threshold by the fluctuation amplitude of the tracking indicators during the historical production process of the gas well specifically refers to statistically analyzing the data fluctuation amplitude of the daily gas production, daily water production, wellhead pressure, flow values ​​at each metering point on the pipeline, and pressure values ​​at each metering point on the pipeline during normal production of the gas well, determining the maximum value of the fluctuation amplitude of different tracking indicators of the gas well under different water-gas ratios, and taking the maximum value of the fluctuation amplitude as the warning threshold.

[0029] Specifically, the actual production data of a certain tracking indicator in the most recent period is selected, and the Laneda criterion is used to eliminate the outliers in the selected actual production data to obtain a valid data set, and the absolute error set of the valid data is obtained based on the valid data set; the maximum absolute error, the second largest absolute error and the third largest absolute error are screened out from the absolute error set of the valid data, and the average of the above three values ​​is calculated as the warning threshold of the tracking indicator.

[0030] Furthermore, the warning threshold is determined by the stabilization time of the tracking indicator using the integrated simulation model. The stabilization time refers to the time when the error between the model prediction value and the actual measurement value of the tracking indicator gradually increases from 0 to the warning threshold. The integrated simulation model is used to perform a sensitivity analysis on the warning threshold of the tracking indicator, and the stabilization time of the tracking indicator is obtained. It is concluded that the water-gas ratio is less than 3m 3 / 10 4 m 3 and greater than 3m 3 / 10 4 m 3 The relationship between the indicator stability time and the warning threshold is calculated, and the warning threshold corresponding to the inflection point in the relationship curve is taken as the threshold of the tracking indicator.

[0031] Determining the tracking frequency in step S103 specifically involves determining the tracking frequency of each tracking indicator according to the data measurement frequency of the actual ground system, the operation time of the integrated simulation model and the tracking frequency requirement.

[0032] Compared with the prior art, the beneficial technical effects brought by the present invention are as follows:

[0033] 1. The present invention can use the predicted values ​​of the integrated simulation model and the real-time information obtained by the ground system sensors for tracking and analysis, and can timely warn and locate the location of model faults, thereby calibrating the model in a targeted manner, ensuring the continuous accuracy of the integrated simulation model, and providing a continuous and accurate model foundation for the construction of intelligent gas fields.

[0034] 2. The present invention determines the early warning threshold by tracking the fluctuation amplitude of the index during the historical production process of the gas well and the stabilization time of the integrated simulation model. These two aspects need to be considered simultaneously when tracking the index. Taking the larger value between the two can make the value of the early warning threshold meet these two aspects at the same time. If the threshold is too small, frequent alarms will occur due to various field data errors or model errors. If the threshold is too large, the sensitivity of the index tracking will be reduced, resulting in partial early warning omissions. In the present invention, the early warning threshold is determined by tracking the fluctuation amplitude of the index during the historical production process of the gas well and the stabilization time of the integrated simulation model, thereby achieving the coordination of early warning accuracy and early warning frequency. The early warning threshold obtained by the method described in the present invention can obtain the optimal value of the early warning threshold.

[0035] 3. In the present invention, the basic starting point for fault location is that the pressure of any node in the system is unique; under stable conditions, the mass of the fluid flowing in and out of each node in the system is conserved. In the present invention, the bottom of the well and the wellhead are taken as solution nodes respectively, and the inflow flow, outflow flow and pressure of the solution node are calculated by an integrated simulation model to determine the true fault location. In this way, the fault location can be quickly located, and the model correction can be carried out on the fault location in a targeted manner to improve the correction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Produce curve fluctuation range charts for tracking indicators;

[0037] Figure 2 The water-gas ratio is less than 3m 3 / 10 4 m 3 The relationship curve between the stable time of the tracking indicator and the warning threshold;

[0038] Figure 3 The water-gas ratio is greater than 3m 3 / 10 4 m 3 The relationship curve between the stable time of the tracking indicator and the warning threshold;

[0039] Figure 4 This is a diagram for wellbore and gas reservoir fault diagnosis. DETAILED DESCRIPTION

[0040] The technical solution of the present invention is further described in detail below in conjunction with specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example 1

[0042] As a preferred embodiment of the present invention, this embodiment discloses a gas reservoir-wellbore-ground integrated simulation model tracking and verification method, comprising the following steps:

[0043] Step S1, using the predicted value of the integrated simulation model to track and compare with the real-time information obtained by the sensors in the actual ground system, and issuing an early warning for the tracking indicators that exceed the early warning threshold;

[0044] Step S2, when an alarm occurs in the production system simulated by the integrated simulation model, a fault location program is started to find the fault location and perform model correction on the fault location; the fault location program includes diagnosing gas reservoir model and / or wellbore model faults with the bottom of the well as the solution node, and diagnosing ground pipeline model faults with the wellhead as the solution node.

[0045] In this embodiment, the integrated simulation model of gas reservoir, wellbore and ground is used to predict the production system, and the predicted value of the integrated simulation model is tracked and compared with the real-time information obtained by the ground system sensors. The specific comparison process is: the actual ground system transmits the real-time information from each sensor in the ground system to the integrated simulation model of gas reservoir, wellbore and ground, and sets the early warning threshold of each tracking index in the integrated simulation model. The integrated simulation model compares the predicted value of each tracking index with the actual value measured by the actual ground system. That is, the absolute error value between the predicted value of the integrated simulation model and the real-time information obtained by the ground system sensors is obtained, and the absolute error value is compared with the preset early warning threshold. If the early warning threshold is exceeded, the production system simulated by the integrated simulation model will alarm.

[0046] As an implementation method of this embodiment, the tracking and comparison described in step S1 is to track the metering data of all ground systems in real time, and the tracking indicators mainly include: daily gas production of the gas well, daily water production of the gas well, wellhead pressure and the flow and pressure of each metering point on the pipeline, such as the downhole pressure gauge. The pressure at the location of the downhole pressure gauge can also be used as a tracking indicator.

[0047] In this embodiment, the warning threshold of the tracking indicator can be preset based on experience.

[0048] In this embodiment, when an early warning appears during the real-time tracking process of the production system, the fault location program is started to find the fault location and perform targeted model correction. The basic starting point for fault location is that the pressure of any node in the system is unique; under stable conditions, the mass of the fluid flowing in and out of each node in the system is conserved. The bottom of the well and the wellhead are taken as the solution nodes respectively, and the inflow flow, outflow flow and pressure of the solution node are calculated through the integrated simulation model to determine the true fault location.

[0049] Example 2

[0050] As another preferred embodiment of the present invention, this embodiment discloses a gas reservoir-wellbore-ground integrated simulation model tracking and verification method, comprising the following steps:

[0051] Step S1: Use the predicted value of the integrated simulation model to track and compare with the real-time information obtained by the sensors in the actual ground system, and issue an early warning for the tracking indicators that exceed the early warning threshold; specifically:

[0052] The integrated simulation model simulates the daily gas production of a fixed gas well, predicts the daily gas production of the gas well, the daily water production of the gas well, the wellhead pressure, the flow value of each metering point on the pipeline, and the pressure value of each metering point on the pipeline;

[0053] The actual ground system transmits the real-time information obtained by each sensor in the ground system to the integrated simulation model; the integrated simulation model tracks and compares the predicted values ​​of each indicator obtained by simulating the daily gas production of a fixed gas well with the real-time information obtained by each sensor in the actual ground system. When the error between a certain tracking indicator and the real-time information exceeds the warning threshold of the tracking indicator, the production system simulated by the integrated simulation model will alarm;

[0054] "Fixed daily gas production of gas wells" means that the integrated model calculates the flow and pressure of all nodes in the model based on the actual daily gas production of the test input. Before the model is operated, the engineer inputs the actual daily gas production into the model, and the model performs simulation operations according to the actual daily gas production input. The "fixed" here means "fixed". When performing simulation, the integrated simulation model can be roughly divided into two types of simulations. One is to calculate the flow and pressure of the entire system with the fixed gas volume mentioned above, and determine how large the gas nozzle size should be to achieve the corresponding gas volume; the other is to fix the gas nozzle size, calculate the flow and pressure of the entire system, and of course the daily gas production of the gas well is also calculated.

[0055] Step S2, when an alarm occurs in the production system simulated by the integrated simulation model, a fault location program is started to find the fault location and perform model correction on the fault location; the fault location program includes diagnosing gas reservoir model and / or wellbore model faults with the bottom of the well as the solution node, and diagnosing ground pipe network model faults with the wellhead as the solution node;

[0056] The method of diagnosing gas reservoir model and / or wellbore model faults by taking the well bottom as the solution node is specifically as follows:

[0057] Using the gas well pipe flow model, the daily gas production Q is calculated by simulating and calculating the wellhead pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. VLP ;

[0058] The daily gas production Q is calculated by using the dynamic characteristic model of gas well inflow and the formation pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. IPR ;

[0059] By analyzing the intersection of the inflow capacity curve and the outflow capacity curve, the gas production Q at the intersection point can be obtained. VLP / IPR ;

[0060] Comparison of daily gas production Q VLP 、Daily gas production Q IPR , gas production at intersection point Q VLP / IPR The actual daily gas production Q obtained from the real-time information obtained by the sensors of the actual ground system T , determine the fault location; if It is determined that the gas reservoir model has a fault;

[0061] like It is determined that the wellbore model has a fault; among them, Er Q is the maximum error allowed for the daily gas production of the integrated simulation model, which is also the warning threshold for the daily gas production.

[0062] The method of diagnosing the fault of the surface pipe network model with the wellhead as the solution node is specifically as follows:

[0063] The actual daily gas production, daily water production and joint station pressure at the wellhead are simulated using the surface pipeline network model to obtain the wellhead pressure P. C ; Compare the wellhead pressure P predicted by the surface pipeline network model C and the actual wellhead pressure P obtained from the real-time information obtained by various sensors of the actual surface system T , determine whether there is a fault in the ground pipe network model. It is determined that there is a fault in the ground pipe network model, where Er P is the maximum error allowed for the wellhead pressure of the integrated simulation model, which is also the warning threshold of the wellhead pressure.

[0064] As an implementation of this embodiment, in step S1, tracking and comparing the predicted value of the integrated simulation model with the real-time information obtained by the sensor in the actual ground system specifically includes:

[0065] Step S101, determining tracking indicators;

[0066] Step S102: determining the warning threshold of each tracking indicator;

[0067] Step S103: Determine the tracking frequency.

[0068] The tracking indicators in step S101 include daily gas production of the gas well, daily water production of the gas well, wellhead pressure, flow values ​​of each metering point on the pipeline, and pressure values ​​of each metering point on the pipeline.

[0069] In step S102, there are two ways to determine the warning threshold of each tracking indicator. The first way is to determine the warning threshold by the fluctuation range of the tracking indicator during the historical production process of the gas well; the second way is to determine the warning threshold by using the stabilization time of the integrated simulation model tracking indicator; the maximum value of the above two ways is taken as the reasonable warning threshold.

[0070] The method of determining the warning threshold by the fluctuation amplitude of the tracking indicators during the historical production process of the gas well specifically refers to statistically analyzing the data fluctuation amplitude of the daily gas production, daily water production, wellhead pressure, flow values ​​at each metering point on the pipeline, and pressure values ​​at each metering point on the pipeline during normal production of the gas well, determining the maximum value of the fluctuation amplitude of different tracking indicators of the gas well under different water-gas ratios, and taking the maximum value of the fluctuation amplitude as the warning threshold.

[0071] Specifically, the actual production data of a certain tracking indicator in the most recent period is selected, and the Laneda criterion is used to eliminate the outliers in the selected actual production data to obtain a valid data set, and the absolute error set of the valid data is obtained based on the valid data set; the maximum absolute error, the second largest absolute error and the third largest absolute error are screened out from the absolute error set of the valid data, and the average of the above three values ​​is calculated as the warning threshold of the tracking indicator.

[0072] Furthermore, the warning threshold is determined by the stabilization time of the tracking indicator using the integrated simulation model. The stabilization time refers to the time when the error between the model prediction value and the actual measurement value of the tracking indicator gradually increases from 0 to the warning threshold. The integrated simulation model is used to perform a sensitivity analysis on the warning threshold of the tracking indicator, and the stabilization time of the tracking indicator is obtained. It is concluded that the water-gas ratio is less than 3m 3 / 10 4 m 3 and greater than 3m 3 / 10 4 m 3 The relationship between the indicator stability time and the warning threshold is calculated, and the warning threshold corresponding to the inflection point in the relationship curve is taken as the threshold of the tracking indicator.

[0073] Determining the tracking frequency in step S103 specifically involves determining the tracking frequency of each tracking indicator based on the data metering frequency of the actual ground system, the operation time of the integrated simulation model, and the tracking frequency requirements. The tracking frequency should be greater than the integrated model operation time and the data metering frequency of the actual ground system. According to actual needs, when the user prefers to track the ground pipeline network, the real-time production data of the SCADA system can be tracked. The data frequency is point-in-second data, and the data volume is rich. The tracking frequency is mainly limited by the model running time, and the tracking frequency is 2-3 times the single model running time. When the user prefers to perform dynamic tracking during gas reservoir analysis, the daily data can be tracked, and the tracking frequency is 1 day / time.

[0074] Example 3

[0075] As another preferred embodiment of the present invention, refer to the attached specification Figure 1-4 This embodiment discloses a gas reservoir-wellbore-ground integrated simulation model tracking and verification method, comprising the following steps:

[0076] Step S1, using the integrated simulation model prediction value and the real-time information obtained by the ground system sensor to track and compare, and issue an early warning for the tracking indicators that exceed the early warning threshold;

[0077] Step S2, when an alarm occurs, start the fault location program, find the fault location, and perform model correction on the fault location; the fault location program includes diagnosing gas reservoir and wellbore model faults with the bottom of the well as the solution node and diagnosing ground pipe network model faults with the wellhead as the solution node;

[0078] In step S1, it is necessary to determine the production system tracking indicators, warning thresholds and tracking frequency; the specific process is as follows:

[0079] (1) Determine tracking indicators

[0080] All on-site metering data are tracked in real time. The tracking indicators mainly include: daily gas production of gas wells, daily water production of gas wells, wellhead pressure, and production and pressure at pipeline metering points. If there is a downhole pressure gauge, the pressure at the downhole pressure gauge location is also taken as a tracking indicator.

[0081] (2) Determine the warning threshold of the tracking indicator

[0082] The warning threshold is a preset value set for the error percentage of each tracking indicator during the real-time tracking and comparison process. When the error percentage between the integrated model prediction value and the actual measurement value exceeds the warning threshold, the production system simulated by the integrated simulation model will alarm. In this embodiment, the warning threshold is determined by the fluctuation amplitude of the indicator in the historical production process of the gas well and the stabilization time of the integrated simulation model.

[0083] 1) Determine the minimum value of the warning threshold by the fluctuation range of the indicator during the historical production process of the gas well

[0084] The index data will experience a certain degree of normal fluctuation in the historical production process. The fluctuation range of the statistical index in normal production in the most recent month (see attached Figure 1 ), set the minimum value of the warning threshold of the tracking indicator to the maximum value of its normal production fluctuation range to prevent frequent false warnings when the indicator is tracked. This case statistically analyzes the data fluctuation range of daily gas production, daily water production, wellhead pressure and pipeline node pressure of 54 gas wells in a gas reservoir in the Sichuan Basin during normal production, and determines the minimum threshold values ​​of different tracking indicators of gas wells under different water-gas ratios. The specific method is as follows:

[0085] ①Select the production data of the indicator in the most recent month: x1, x2, ..., x 30

[0086] ② Use the Laneda criterion to eliminate outliers

[0087] Find the arithmetic mean:

[0088]

[0089] Find the absolute error:

[0090]

[0091] Find the RMS deviation:

[0092]

[0093] Outlier judgment: Then x i After removing the outliers, there are n (1≤n≤30) valid data remaining, which are expressed as a set: X= <x1,x2,……,x n >.

[0094] ③The minimum value of the warning threshold is the average of the three values ​​with the largest absolute errors of the valid data.

[0095] Find the absolute error of the valid data:

[0096]

[0097] The absolute error of the valid data is expressed as a set: ΔX = < Δx1, Δx2, ..., Δx n >

[0098] Maximum absolute error: Y1 = max(ΔX)

[0099] The second largest absolute error: Y2 = max(ΔX-Y1)

[0100] The third largest value of absolute error: Y3 = max(ΔX-Y1-Y2)

[0101] The minimum value of the warning threshold is:

[0102]

[0103] ④ Repeat the above steps until the minimum warning threshold value of all tracking indicators is calculated.

[0104] In practical applications, the minimum value of the warning threshold should also be determined based on the actual situation of the mine. If there is less historical production data for the mine, the threshold in Table 1 below can also be used for reference.

[0105]

[0106] Table 1 shows the minimum warning thresholds of different tracking indicators at different water-gas ratios.

[0107] 2) Determine the warning threshold using the stability time of the integrated simulation model indicators

[0108] The accuracy of the integrated simulation model will gradually change with the development and production of the gas field. The stability time of the integrated simulation model index is defined as the time when the error between the model prediction value and the actual measurement value of the index gradually increases from 0 to the threshold. The larger the threshold, the longer the stability time of the index, and the lower the frequency of model correction; the smaller the threshold, the shorter the stability time of the index, and the higher the frequency of model correction.

[0109] In this case, an integrated simulation model is used to conduct sensitivity analysis on the indicator warning threshold. The indicator warning threshold range is 2-16%, the interval is 2%, and the sensitivity analysis is performed 8 times. The model defines different warning thresholds for simulation, and the stability time of the predicted indicator is obtained. It is concluded that the water-gas ratio is less than 3m 3 / 10 4 m 3 and greater than 3m 3 / 10 4 m 3 The relationship between the indicator stability time and threshold value (see Appendix Figure 2 , Attachment Figure 3 ), and take its inflection point as the reasonable warning threshold.

[0110] The larger value of the above two methods is taken as the reasonable warning threshold. See Table 2 below.

[0111]

[0112] Table 2 shows the reasonable warning thresholds for different tracking indicators.

[0113] (3) Determine the tracking frequency

[0114] The tracking frequency is mainly determined by the field data metering frequency, the integrated model operation time and the tracking frequency requirements. The tracking frequency should be greater than the integrated model operation time and the field data metering frequency. According to actual needs, when users prefer to track the surface pipeline network, they can track the real-time production data of the SCADA system. The data frequency is point-in-second data with rich data volume. The tracking frequency is mainly limited by the model running time. The tracking frequency is 2-3 times the single model running time. When users prefer dynamic tracking during gas reservoir analysis, they can track daily data, and the tracking frequency is 1 day / time.

[0115] Among them, in step S2, the fault location procedure includes diagnosing gas reservoir and wellbore model faults with the bottom of the well as the solution node and diagnosing ground pipe network model faults with the wellhead as the solution node. Specifically, it means:

[0116] When an early warning appears during the real-time tracking of the production system, the fault location program is started to find the fault location and perform targeted model correction. The basic starting point for fault location is that the pressure of any node in the system is unique; under stable conditions, the mass of the fluid flowing in and out of each node in the system is conserved. The bottom of the well and the wellhead are taken as the solution nodes respectively, and the inflow flow, outflow flow and pressure of the solution node are calculated through the integrated simulation model to determine the true fault location.

[0117] ① Diagnose gas reservoir and wellbore model faults using the bottom of the well as the solution point

[0118] Using the gas well pipe flow model, the daily gas production is calculated by the wellhead pressure and bottom hole flow pressure, denoted as Q VLP ; Using the gas well inflow dynamic characteristic model, the daily gas production is calculated by the formation pressure and bottom hole flow pressure, denoted as Q IPR ; Through the intersection analysis of the inflow capacity curve and the outflow curve, the output at the intersection is recorded as Q VLP / IPR ; The actual daily gas production is recorded as Qt. By comparing Q VLP , Q IPR , Q VLP / IPR , Q T Determine the fault location and the judgment method is as follows:

[0119] if Then there is a fault in the gas reservoir model;

[0120] if There is a fault in the wellbore model.

[0121] Among them, Q is the maximum error allowed for the daily gas production of the model, and this value is the warning threshold of the daily gas production.

[0122] ② Diagnose surface pipe network model faults using the wellhead as the solution point

[0123] Since the wellhead daily gas production, daily water production, wellhead pressure and wellhead transmission pressure are usually metering data, the surface pipeline network model can be directly used to determine the actual wellhead daily gas production, daily water production and joint station pressure to calculate the wellhead transmission pressure P C , the actual input pressure is recorded as P T By comparing P C , P T Determine whether there is a fault in the ground pipe network. The judgment method is as follows:

[0124] if Then the ground pipe network model has a fault, where Er P is the maximum error allowed for the wellhead pressure of the integrated simulation model, which is also the warning threshold of the wellhead pressure.

[0125] According to the fault location of the integrated simulation model, the model is corrected in a targeted manner to ensure the accuracy of the model. The fitting correction method for the gas reservoir model can be found in the "Technical Specifications for the Application of Numerical Simulation of Oil and Gas Reservoirs", standard number SY / T 6744-2008; the high-frequency fitting method for the wellbore and IPR can be found in the "Correction Method, Device and Computer Reservoir Medium for the Gas Well Inflow Dynamic Curve", patent number 201910716909.X, and the detailed correction process will not be repeated in this invention.

[0126] Example 4

[0127] Taking a gas reservoir in Sichuan as an example, the field metering parameters of the gas reservoir mainly include the daily gas production of the gas well, the daily water production, the wellhead pressure and the pipeline metering pressure, which are stored in the database. The gas reservoir has adopted the IPM software of PETEX to complete the construction of the gas reservoir-wellbore-ground integrated simulation model. On this basis, the production system is tracked, warned and faults are located, guiding the verification of the integrated simulation model to ensure the continuous accuracy of the model.

[0128] The tracking indicators determined based on the on-site metering data of the gas reservoir include: daily gas production of the gas well, daily water production, wellhead pressure and pressure at the pipeline metering point.

[0129] A program was written in Visual Basic to calculate the threshold of the tracking indicator. First, the data of the tracking indicator during normal production in the most recent month was extracted from the database, and the minimum threshold of the corresponding indicator was calculated using the formula in the technical solution of the present invention (see Table 1 above). Then, the GAP module in the IPM software was opened, and the integrated gas reservoir simulation model was opened to automatically perform a one-year production forecast operation. The daily gas production, daily water production, wellhead pressure, and pressure at the pipeline metering point calculated by the model were extracted and compared with the on-site metering values. The time when the indicator error was greater than the threshold was recorded as the indicator stabilization time. By changing the threshold and repeating the calculation and comparison, the relationship between the indicator stabilization time and the threshold can be obtained (see Appendix Figure 1 and attached Figure 2 ), and take its inflection point as the tracking threshold.

[0130] The larger value of the above two methods is taken as the reasonable threshold (see Table 3).

[0131]

[0132] Table 3 is a production system tracking and early warning information table. This embodiment tracks the daily data in the database and determines the tracking frequency as daily tracking.

[0133] The program code "ZBYC" is written in Visual Basic to realize the functions of automatic opening of the integrated model, automatic data entry, automatic model calculation, and automatic storage of calculation results in the database. Through this program, the GAP module in the IPM software is opened, and the integrated simulation model of the gas reservoir that has been established is opened, and the Opensever command (IPM software external data dedicated interface, which can connect external data to the software for use) is called to automatically enter the daily gas production and separator pressure in the database into the integrated simulation model. After the entry is completed, the integrated simulation model performs calculations to obtain the daily gas production, daily water production, wellhead pressure, pipeline node flow and pressure of the gas well, and store them in the prediction index table in the database. The "ZBYC" program is set to be automatically executed once every morning at zero o'clock, and the index prediction value is automatically appended to the prediction index table.

[0134] The program code "GZYJ" is written in Visual Basic. The program automatically extracts the predicted index value of the prediction index table in the database and compares it with the measured value in the database. When the index error of a node exceeds the threshold, the corresponding node and index are highlighted and flashed on the ground pipe network map for early warning, and stored in the early warning information table in the database (see Table 3). It is set that after the execution of the "ZBYC" program code is completed every day, the "GZYJ" program code is executed to ensure that the production system performs a model failure early warning once a day.

[0135] The program code "GZDW" was written in Visual Basic. This code automatically locates the real fault location that causes the warning for the model fault warning that occurs during the tracking warning process, and stores it in the fault location column in the warning information table (see Table 3). The program first opens the IPR module of the PROSPER software, automatically inputs the formation pressure and the test well bottom flow pressure in the integrated simulation model, and then calculates Q IPR ; Open the VLP module, input the measured wellhead pressure and bottom hole flow pressure, and perform the calculation to get Q VLP ; Open the system analysis module and perform calculations to obtain Q VLP / IPR ; The actual daily gas production is recorded as Qt. By comparing Q VLP , QIPR , Q VLP / IPR , Qt determine the fault location of the gas reservoir and wellbore model. And draw the wellbore and gas reservoir fault location diagnosis map (see Figure 2 ) Assist engineers to conduct fault analysis. The diagram includes wellbore string diagram, IPR and VLP curves, and production calculated by IPR method, VLP method, and IPR / VLP curve intersection method. The program closes PROSPER software, opens GAP software and opens the ground pipeline network model, and automatically sets the ground pipeline network model to use the fixed-meter wellhead daily gas production, daily water production, and joint station pressure for calculation. After the calculation, the wellhead pressure PC is obtained, and the metered pressure is recorded as PT. By comparing PC and PT, it is determined whether there is a fault in the ground pipeline network. After each "GZYJ" program code is executed, the early warning information table is monitored. If an early warning message appears, the "GZDW" program code is automatically executed.

[0136] According to the fault position of the integrated simulation model located, a targeted correction model is used. For the high-frequency fitting method of the wellbore and IPR, please refer to "Correction method, device and computer reservoir medium for gas well inflow dynamic curve", patent number 201910716909.X, to ensure the continuous accuracy of the integrated simulation model. The detailed correction process will not be repeated in this invention.

Claims

1. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method, characterized in that: The following steps are involved: Step S1, using the predicted value of the integrated simulation model to track and compare with the real-time information obtained by the sensors in the actual ground system, and issuing an early warning for the tracking indicators that exceed the early warning threshold; Step S2, when an alarm occurs in the production system simulated by the integrated simulation model, a fault location program is started to find the fault location and perform model correction on the fault location; the fault location program includes diagnosing gas reservoir model and / or wellbore model faults with the bottom of the well as the solution node, and diagnosing ground pipe network model faults with the wellhead as the solution node; The method of diagnosing gas reservoir model and / or wellbore model faults by taking the well bottom as the solution node is specifically as follows: Using the gas well pipe flow model, the daily gas production Q is calculated by simulating and calculating the wellhead pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. VLP ; The daily gas production Q is calculated by using the dynamic characteristic model of gas well inflow and the formation pressure, bottom hole flow rate and bottom hole pressure predicted by the integrated simulation model according to the daily gas production of the fixed gas well. IPR ; By analyzing the intersection of the inflow capacity curve and the outflow capacity curve, the gas production Q at the intersection point can be obtained. VLP / IPR ; Comparison of daily gas production Q VLP 、Daily gas production Q IPR , gas production at intersection point Q VLP / IPR The actual daily gas production Q obtained from the real-time information obtained by the sensors of the actual ground system T , determine the fault location; The method of diagnosing the fault of the surface pipe network model with the wellhead as the solution node is as follows: the actual wellhead daily gas production, daily water production and joint station pressure are determined by the surface pipe network model to perform simulation calculations to obtain the wellhead transmission pressure P C ; Compare the wellhead pressure P predicted by the surface pipeline network model C and the actual wellhead pressure P obtained from the real-time information obtained by various sensors of the actual surface system T , determine whether there is any fault in the ground pipe network model.

2. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 1, characterized in that: The step S1 is specifically as follows: The integrated simulation model simulates the daily gas production of a fixed gas well, predicts the daily gas production of the gas well, the daily water production of the gas well, the wellhead pressure, the flow value of each metering point on the pipeline, and the pressure value of each metering point on the pipeline; The actual ground system transmits the real-time information obtained by each sensor in the ground system to the integrated simulation model; The integrated simulation model tracks and compares the predicted values ​​of various indicators obtained based on the daily gas production simulation of the fixed gas well with the real-time information obtained by each sensor in the actual ground system. When the error between a certain tracking indicator and the real-time information exceeds the warning threshold of the tracking indicator, the production system simulated by the integrated simulation model will alarm.

3. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 1, characterized in that: like , it is determined that the gas reservoir model has a fault; like , it is determined that the wellbore model has a fault; among them, is the maximum error allowed for the daily gas production of the integrated simulation model, which is also the warning threshold for the daily gas production.

4. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 1, characterized in that: like , it is determined that there is a fault in the ground pipe network model, where is the maximum error allowed for the wellhead pressure of the integrated simulation model, which is also the warning threshold of the wellhead pressure.

5. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method as claimed in claim 1 or 2, characterized in that: In step S1, the prediction value of the integrated simulation model is tracked and compared with the real-time information obtained by the sensors in the actual ground system, which specifically includes: Step S101, determining tracking indicators; Step S102: determining the warning threshold of each tracking indicator; Step S103: Determine the tracking frequency.

6. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method as claimed in claim 5, characterized in that: The tracking indicators in step S101 include daily gas production of the gas well, daily water production of the gas well, wellhead pressure, flow values ​​of each metering point on the pipeline, and pressure values ​​of each metering point on the pipeline.

7. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method as claimed in claim 5, characterized in that: In step S102, there are two ways to determine the warning threshold of each tracking indicator. The first way is to determine the warning threshold by the fluctuation amplitude of the tracking indicator during the historical production process of the gas well; the second way is to determine the warning threshold by using the stabilization time of the integrated simulation model tracking indicator; the maximum value of the above two ways is taken as the reasonable warning threshold.

8. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 7, characterized in that: Determining the warning threshold by the fluctuation amplitude of the tracking indicators during the historical production process of the gas well specifically refers to statistically analyzing the data fluctuation amplitude of the daily gas production, daily water production, wellhead pressure, flow values ​​at each metering point on the pipeline, and pressure values ​​at each metering point on the pipeline during normal production of the gas well, determining the maximum value of the fluctuation amplitude of different tracking indicators of the gas well under different water-gas ratios, and taking the maximum value of the fluctuation amplitude as the warning threshold.

9. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method as claimed in claim 8, characterized in that: The actual production data of a certain tracking indicator in the most recent period is selected, and the outliers in the selected actual production data are eliminated using the Laneda criterion to obtain a valid data set. The absolute error set of the valid data is obtained based on the valid data set; the maximum absolute error, the second largest absolute error and the third largest absolute error are screened out from the absolute error set of the valid data, and the average of the above three values ​​is calculated as the warning threshold of the tracking indicator.

10. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 7, characterized in that: The integrated simulation model is used to track the stability time of the indicator to determine the warning threshold, and the stability time refers to the time when the error between the model prediction value and the actual measurement value of the tracking indicator gradually increases from 0 to the warning threshold; An integrated simulation model is used to conduct sensitivity analysis on the warning threshold of the tracking indicator, and the stability time of the tracking indicator is obtained. It is concluded that the water-gas ratio is less than 3m 3 / 10 4 m 3 and greater than 3m 3 / 10 4 m 3 The relationship between the indicator stability time and the warning threshold is calculated, and the warning threshold corresponding to the inflection point in the relationship curve is taken as the threshold of the tracking indicator.

11. A gas reservoir-wellbore-ground integrated simulation model tracking and verification method according to claim 5, characterized in that: Determining the tracking frequency in step S103 specifically involves determining the tracking frequency of each tracking indicator according to the data measurement frequency of the actual ground system, the operation time of the integrated simulation model and the tracking frequency requirement.

Images