Well wall stability simulation evaluation device and method under fault activation condition

Abstract

The invention relates to a well wall stability simulation evaluation device and method under a fault activation condition. The device comprises a test core, a simulation well wall, a simulation wellhead, a confining pressure application and temperature control application module, a manifold control module and a central control module. The simulated well wall is a simulated well wall which is drilled inwards from the wall surface of the test core and is used for simulating an open hole well wall which is not subjected to well cementation; the central control module controls the confining pressure applying and temperature control applying module and the manifold control module to change confining pressure and temperature and simulate fluid pressure in a well wall, and fault activation conditions in all directions and the influence of the fault activation conditions on the stability of the well wall are judged. Compared with the prior art, the device can simulate the stability and instability process of the well wall under the combined action of natural crack weak surface activation, well periphery stress field change, high-pressure drilling fluid soaking and the like, explores the influence mechanism and control method of the stability of the well wall, and provides a solid guarantee for on-site safe drilling.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of petroleum engineering, in particular to a wellbore stability simulation evaluation device and method under fault activation condition. BACKGROUND

[0002] Shale gas has developed micro-fractures, which are easy to break when encountering drilling fluid. During drilling, especially during shallow drilling, wellbore stability is one of the important factors to ensure the safety and efficiency of the drilling process. Due to the complex and changeable downhole environment, the wellbore rock is subjected to the change of drilling-induced stress field and is also soaked by high-pressure drilling fluid. In this process, the weak plane of the natural fracture system may be activated (i.e. fault activation), which may further cause wellbore instability risk. Exploring the influence mechanism and control method of natural weak plane on wellbore stability is an important research content for safe drilling in shallow complex fracture network, which provides an important guarantee for high-quality and safe drilling of wellbore.

[0003] At present, in addition to using field logging data, the research on wellbore stability usually involves making sample blocks and adding true triaxial or pseudo triaxial confining pressure to manufacture artificial wellbore in the laboratory to determine the wellbore stability condition. However, for complex fracture systems with natural weak planes, and under the condition that the wellbore is soaked by high-pressure drilling fluid, the positional relationship between the wellbore and the natural fractures, the change sequence of the surrounding stress, and the change of the drilling fluid properties will all affect the wellbore stability. The commonly used experimental device cannot simulate the wellbore stability under the condition of fault activation. Therefore, it is necessary to develop a wellbore stability simulation evaluation device to explore the influence mechanism and control method of wellbore stability under the combined action of weak plane, drilling fluid and stress state, and to provide a solid guarantee for safe drilling on site.

[0004] Patent CN201610225445.9 discloses a fracture-matrix coupling flow damage evaluation device and method simulating formation conditions. The device mainly consists of a kettle body, a core, a fracture, a simulated borehole, a pressure-increasing capsule, a resistance probe, a pressure sensor, a data acquisition system, a working fluid circulation system, etc. The kettle body is provided with an axial pressure injection inlet, a confining pressure injection inlet and an air inlet, and the core is surrounded by the pressure-increasing capsule. The core has a fracture and a simulated borehole, and is provided with nine drill holes penetrating the core, and the drill holes are provided with resistance probes. The resistance probes are connected with the pressure sensor, and the resistance probes and the pressure sensor are connected with the data acquisition system. The simulated borehole is connected with the working fluid circulation system. The device is used to simulate the wellbore secondary stress condition of the fracture-matrix system and the formation radial flow to determine the damage degree of the fracture-matrix system. However, the device and method are relatively complex, and are not based on the formation of a database through experiments and then input into a model for training to obtain a mature simulation model. Therefore, the evaluation device and method have limitations and are not accurate. SUMMARY

[0005] The present application aims to overcome the defects of the prior art and provide a wellbore stability simulation evaluation device and method under fault activation conditions, which can simulate the stability and instability process of the wellbore under the combined action of natural fracture weak surface activation, wellbore stress field change and high-pressure drilling fluid soaking, and explore the influencing mechanism and control method of wellbore stability, thereby providing a solid guarantee for safe drilling on site.

[0006] The object of the present application can be achieved by the following technical solutions:

[0007] The present application provides a wellbore stability simulation evaluation device under fault activation conditions, comprising: a test core, a simulated wellbore, a simulated wellhead, a confining pressure application and temperature control application module, a manifold control module, and a central control module.

[0008] The test core is internally provided with a natural weak surface or an artificial weak surface.

[0009] The simulated wellbore is a cylindrical simulated wellbore drilled from the wall surface of the test core inward, used to simulate the open hole wellbore without cementing, and the opening of the simulated wellbore comprises at least three mutually perpendicular surfaces of the simulated wellbore, used to simulate different drilling directions underground, and each surface of the simulated wellbore comprises at least two wells that are perpendicular and inclined to the wall surface of the test core, and each simulated wellbore intersects with a fault, used for comparative analysis of the stability mechanism.

[0010] The simulated wellhead is installed at the opening of the simulated wellbore on the wall surface of the test core, one end of the simulated wellhead is connected with the manifold control module through a pressure manifold, and the manifold control module can inject high-pressure fluid into each simulated wellbore and maintain the pressure.

[0011] The confining pressure application and temperature control application module is arranged outside the test core, and the confining pressure application and temperature control application module can apply confining pressure to the outer wall surface of the test core and can heat the test core and maintain the temperature of the test core constant.

[0012] The central control module is in communication connection with the confining pressure application and temperature control application module and the manifold control module, and the central control module changes the confining pressure, temperature and fluid pressure in the simulated wellbore by controlling the confining pressure application and temperature control application module and the manifold control module, judges the fault activation in each direction and its influence on the wellbore stability, and thereby guides the analysis of the influencing mechanism and control method of the fault activation on the wellbore stability.

[0013] Further, the test core is a natural outcrop or a man-made core.

[0014] Further, the plurality of simulated wellbores cannot penetrate or directly penetrate the test core.

[0015] Further, the simulated wellhead is in sealing connection with the simulated wellbore and the pressure manifold, respectively.

[0016] Further, the confining pressure applying and temperature control applying module can apply true tri-axial pressure to the outer wall of the test core.

[0017] Further, the central control module can control the sequence, size and time of the liquid pressure applied to each simulated well wall by the manifold control module.

[0018] Further, the central control module can control the size, sequence and time of the confining pressure applied to the test core by the confining pressure applying and temperature control applying module.

[0019] Further, the central control module can control the temperature and constant temperature control applied to the test core by the confining pressure applying and temperature control applying module.

[0020] The application also provides a method for simulating and evaluating the well wall stability under fault activation conditions, comprising the following steps:

[0021] S1, selecting outcrop or manufacturing artificial core of the target reservoir rock sample, processing into square test core for standby, the test core containing natural weak plane or artificial weak plane;

[0022] S2, drilling holes on each surface of the test core to manufacture simulated well walls, the opening of the simulated well wall containing at least three mutually perpendicular surfaces of the simulated well wall, to simulate different drilling azimuths under the well, and each surface containing at least two wells vertical and inclined to the wall surface of the test core, and each simulated well wall intersecting with the fault, and taking photos of the internal micro-crack state of each simulated well wall by imaging means;

[0023] S3, installing a simulated wellhead on each simulated well wall and connecting the manifold control module through the pressure manifold;

[0024] S4, placing the test core inside the confining pressure applying and temperature control applying module;

[0025] S5, controlling the confining pressure applying and temperature control applying module to heat the test core to reach the set constant temperature state by the central control module, to simulate the formation condition;

[0026] S6, writing the size, sequence and duration instructions of the confining pressure applying and fracturing fluid applying of each simulated wellhead in the central control module;

[0027] S7, starting the experiment, the confining pressure applying and temperature control applying module applying confining pressure to the test core to simulate the formation stress state, and the manifold control module injecting liquid into each simulated well wall and applying pressure to simulate the open hole well wall in the drilling process;

[0028] S8, after the instruction in the central control module is applied, the confining pressure and the fluid pressure in the wellbore are simultaneously removed slowly to prevent stress mutation from causing secondary damage to the simulated wellbore, and the test core is taken out;

[0029] S9, the imaging means is used to take pictures of the state of the microcracks in each simulated wellbore again, the activation state of the natural weak plane and the microcracks is determined according to the displacement of the microcracks on the simulated wellbore, the stability of the simulated wellbore is determined according to the broken and missing parts of the simulated wellbore, and the state of the microcracks on the simulated wellbore is recorded, and the state of the microcracks on the simulated wellbore is divided into four categories: stable wellbore and no fault activation, unstable wellbore and tensile fracture of fault activation, unstable wellbore and shear fracture of fault activation, and unstable wellbore and tensile-shear fracture of fault activation; the determination method is that, based on the observation results of the imaging means, if the fault at the natural weak plane or the artificial weak plane does not shear or open, it is "stable wellbore and no fault activation"; if the fault at the natural weak plane or the artificial weak plane shears, it is "unstable wellbore and shear fracture of fault activation"; if the fault at the natural weak plane or the artificial weak plane opens, it is "unstable wellbore and tensile fracture of fault activation"; and if the fault at the natural weak plane or the artificial weak plane shears and opens, it is "unstable wellbore and tensile-shear fracture of fault activation";

[0030] S10, a large amount of data is obtained by repeating the experiment, principal component analysis is performed according to the data of the experiment, weight coefficients are obtained, and the influencing factors and mechanism of the wellbore stability are analyzed through the weight coefficients; then, a relationship model between the wellbore orientation, the natural weak plane, the stress state, the temperature, the wellbore pressure and the fault activation / natural weak plane activation state and the wellbore stability is established in the central control module, the influence mechanism of the fault activation on the wellbore stability is analyzed, and the control method for avoiding the fault activation / natural weak plane activation under the drilling condition is analyzed.

[0031] Further, in S10, a group of data is recorded at the place with the natural crack on each wellbore in each experiment, and the group of data is taken as a row in the data table, wherein the input parameters of the group of data are the wellbore orientation at the intersection of the natural crack and the wellbore, the angle between the natural weak plane and the wellbore, the stress state, the temperature and the wellbore pressure, and the output parameters are the activation state of the natural weak plane and the microcracks and the wellbore stability; different natural cracks on different wellbores in different experiments are counted as different rows in the data set;

[0032] The wellbore orientation and the angle between the natural weak plane and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak plane; the stress state, the temperature and the wellbore pressure are obtained through the sensors arranged in the test core;

[0033] The data set is obtained by continuously changing the parameters and repeating the experiment;

[0034] The dataset was processed by normalizing the wellbore orientation, the angle between the natural weak surface and the wellbore, the stress state, temperature, and wellbore pressure parameters to between 0 and 1. The activation state of the natural weak surface and microfractures, and the wellbore stability were classified according to the previous step: "wellbore stable and without fault activation" were assigned a weight of 0; "wellbore unstable and with fault activation (shear fracture)" were assigned a weight of 1; "wellbore unstable and with fault activation (open fracture)" were assigned a weight of 2; and "wellbore unstable and with fault activation (tensile-shear fracture)" were assigned a weight of 3. This yielded the processed dataset.

[0035] Principal component analysis (PCA) was used to analyze the weights of the input parameters, obtaining weight coefficients. These coefficients were then sorted from largest to smallest; the larger the weight, the more significant the influence of the parameter on fault activation and natural weak surface activation. The weight coefficients are a quantitative representation of wellbore stability under fault activation conditions, and the influencing factors and mechanisms of wellbore stability were analyzed using these weight coefficients.

[0036] Furthermore, in S10,

[0037] Establish a machine learning classification model to predict the activation state of natural weak surfaces and microfractures, as well as wellbore stability;

[0038] During model building, a set of data was recorded for each location with natural cracks on the wellbore in each experiment, and this data was recorded as a row in the data table. The input parameters for this set of data were the wellbore orientation at the intersection of the natural crack and the wellbore, the angle between the natural weak surface and the wellbore, the stress state, the temperature, and the wellbore pressure. The output parameters were the activation state of the natural weak surface and the microcracks, and the wellbore stability. Different natural cracks on different wellbores in different experiments were counted as different rows in the dataset.

[0039] The wellbore orientation and the angle between the natural weak surface and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak surface; the stress state, temperature, and wellbore pressure are obtained through sensors arranged inside the test core.

[0040] By repeatedly experimenting with constantly changing parameters, a dataset can be obtained.

[0041] In the process of establishing the machine learning classification model, the dataset is used as the training model. The input parameters are the wellbore orientation at the intersection of natural fractures and wellbore, the angle between the natural weak surface and the wellbore, the stress state, the temperature, and the wellbore pressure. After normalization, the output parameters are the activation state of the natural weak surface and microfractures, and the wellbore stability. They are also divided into four categories and labeled with weights. The input parameters are used as the sample dataset, and the corresponding output parameters are used as the label dataset. Machine learning is then performed to obtain the trained machine learning classification model.

[0042] When the machine learning classification model is applied, a real-time evaluation of the wellbore stability under the fault activation condition is realized, that is, based on the field logging interpretation data, on the target layer with the same rock as the experiment, the intersection of the natural fracture and the wellbore direction, the angle between the natural weak plane and the wellbore, the stress state, the temperature, the wellbore pressure and other parameters on the logging interpretation curve are normalized in the same way as in the experiment, and are used as input parameters, and the activation state of the natural weak plane and the micro fracture and the prediction weight of the wellbore stability are obtained through the trained machine learning classification model, so that the prediction of whether the natural weak plane and the micro fracture are activated and whether the wellbore is stable in the real wellbore is realized.

[0043] The fault activation mechanism is analyzed through the prediction model.

[0044] Further, in S2 and S9, the imaging means includes a camera, a small wellbore imaging system, a CT and a nuclear magnetic.

[0045] The design principle of the present application is:

[0046] In the laboratory, samples containing natural weak planes and micro fractures are constructed to manufacture simulated wellbores of different directions and shapes, different confining pressures and temperatures are applied on the test core, while the drilling fluid pressure is applied inside the simulated wellbore, so as to reproduce the environment encountered by the wellbore in the drilling process of shallow shale, and then the activation of the micro fracture and the natural weak plane in each simulated wellbore is counted, and the stability of the simulated wellbore after the fault activation / natural weak plane activation is counted, a large amount of data is obtained through a large number of repeated experiments, a database of fault activation / natural weak plane activation conditions under the conditions of formation temperature, stress and fracturing fluid soaking is constructed, and a database of wellbore stability under the conditions of fault activation / natural weak plane activation is constructed, which is used for further analysis of the influence mechanism of fault activation on wellbore stability.

[0047] Compared with the prior art, the present application has the following advantages:

[0048] The present application has simple and easy-to-understand experimental steps, convenient operation, low test cost, high test speed, can obtain a large amount of data through a large number of repeated experiments, and can construct a database of fault activation / natural weak surface activation conditions under the conditions of formation temperature, stress and drilling fluid soaking, and a database of whether the wellbore is stable under the conditions of fault activation / natural weak surface activation, understand the correlation and influence degree between different variables by using principal component analysis method, and achieve the purpose of analyzing the influence mechanism and control method of fault activation on wellbore stability, which has a positive effect on shale gas drilling risk evaluation and drilling design. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 It is a schematic diagram of the wellbore stability simulation and evaluation device under the condition of fault activation.

[0050] The figure mark: 1, test core; 2, simulated wellbore; 3, simulated wellhead; 4, confining pressure application and temperature control application module; 5, manifold control module; 6, central control module. DETAILED DESCRIPTION

[0051] The present application will be described in detail below in combination with the drawings and specific embodiments. In the technical solution, the component model, material name, connection structure, control method, algorithm and other features not explicitly described are considered as common technical features disclosed in the prior art.

[0052] Example 1

[0053] The present application provides a wellbore stability simulation and evaluation device under the condition of fault activation, which comprises: a test core 1, a simulated wellbore 2, a simulated wellhead 3, a confining pressure application and temperature control application module 4, a manifold control module 5, and a central control module 6, wherein the central control module 6 is a computer.

[0054] The test core 1 is internally provided with a natural weak surface or an artificial weak surface;

[0055] The simulated wellbore 2 is a cylindrical simulated wellbore drilled inward from the wall surface of the test core 1, which is used to simulate the open hole wellbore without cementing. The opening of the simulated wellbore 2 comprises at least three mutually perpendicular surfaces of the simulated wellbore 2, which is used to simulate different drilling directions in the wellbore, and each surface of the simulated wellbore 2 comprises at least two wells vertical and inclined to the wall surface of the test core 1, and each simulated wellbore 2 intersects with a fault, which is used for comparative analysis of the stability mechanism.

[0056] The simulated wellhead 3 is installed at the opening of the simulated wellbore 2 on the wall surface of the test core 1, one end of the simulated wellhead 3 is connected with the manifold control module 5 through a pressure manifold, and the manifold control module 5 can inject high-pressure fluid into each simulated wellbore 2 and maintain the pressure;

[0057] The confining pressure applying and temperature controlling module 4 is arranged outside the test core 1, and can apply confining pressure to the outer wall surface of the test core 1, and can heat the test core 1 and keep the temperature of the test core 1 constant.

[0058] The central control module 6 is in communication connection with the confining pressure applying and temperature controlling module 4 and the manifold control module 5 respectively, and the central control module 6 changes the confining pressure, temperature and fluid pressure in the simulated well wall 2 by controlling the confining pressure applying and temperature controlling module 4 and the manifold control module 5, judges the fault activation in each direction and the influence on the well wall stability, so as to guide the analysis of the influence mechanism and control method of the fault activation on the well wall stability.

[0059] In the specific embodiment, the confining pressure applying and temperature controlling module 4 is composed of a true triaxial compression machine and a temperature sensor, the true triaxial compression machine can apply pressure in two horizontal directions and one vertical direction to the rock sample, and the temperature sensor is located inside the true triaxial compression machine and can monitor the surface temperature of the sample in real time. The manifold control module 5 is mainly composed of a fracturing fluid tank, a high-pressure pump and a valve control. The central control module 6 can control the opening and closing state of the valve of each pipeline output by the manifold control module 5, and can also control the start and stop state of the high-pressure pump. There is a pressure gauge on the pipeline, which can feed back signals to the central control module 6.

[0060] In the specific embodiment, the test core 1 is a natural outcrop or a man-made core.

[0061] In the specific embodiment, the plurality of simulated well walls 2 cannot penetrate or directly penetrate the test core 1.

[0062] In the specific embodiment, the simulated wellhead 3 is sealingly connected with the simulated well wall 2 and the pressure manifold.

[0063] In the specific embodiment, the confining pressure applying and temperature controlling module 4 can apply true triaxial pressure to the outer wall surface of the test core 1.

[0064] In the specific embodiment, the central control module 6 can control the sequence, size and time of the liquid pressure applied by the manifold control module 5 to each simulated well wall 2.

[0065] In the specific embodiment, the central control module 6 can control the size, sequence and time of the confining pressure applied by the confining pressure applying and temperature controlling module 4 to the test core.

[0066] In the specific embodiment, the central control module 6 can control the temperature and constant temperature control applied by the confining pressure applying and temperature controlling module 4 to the test core.

[0067] The embodiment also provides a well wall stability simulation evaluation method under fault activation conditions, comprising the following steps:

[0068] S1, select outcrop or artificial core of the target reservoir rock sample, process into square test core 1, which contains natural or artificial weak surface;

[0069] S2, drill holes on each surface of the test core 1 to make simulated wellbore 2, which contains at least three perpendicular surfaces of the simulated wellbore 2 at the opening, to simulate different drilling azimuths in the well, and each surface contains at least two wells that are perpendicular and inclined to the wall surface of the test core 1, and each simulated wellbore 2 intersects with a fault, and the internal micro-fracture state of each simulated wellbore 2 is photographed and archived by imaging means;

[0070] S3, install simulated wellhead 3 on each simulated wellbore 2 and connect tubing control module 5 through the pressure manifold;

[0071] S4, place the test core 1 into the confining pressure and temperature control module 4;

[0072] S5, control the confining pressure and temperature control module 4 to heat the test core 1 to a constant temperature state set by the central control module 6, to simulate the formation conditions;

[0073] S6, write the size, sequence and duration of confining pressure and fracturing fluid application instructions for each simulated wellhead 3 in the central control module 6;

[0074] S7, start the experiment, the confining pressure and temperature control module 4 applies confining pressure to the test core to simulate the formation stress state, and the tubing control module 5 injects fluid into each simulated wellbore 2 and applies pressure to simulate the open hole wall during the drilling process;

[0075] S8, after the instructions in the central control module 6 are applied, simultaneously slowly remove the confining pressure and fluid pressure in the wellbore to prevent stress mutation from causing secondary damage to the simulated wellbore 2, and remove the test core 1;

[0076] S9, re-photographing and archiving the internal micro-fracture state of each simulated borehole wall 2 by imaging means, judging the activation state of natural weak plane and micro-fracture through the displacement of the micro-fracture on the simulated borehole wall 2, judging the stability of the simulated borehole wall 2 and recording through the broken and missing of the simulated borehole wall 2, and dividing the micro-fracture damage state on the simulated borehole wall 2 into four categories: stable borehole wall without fault activation, tensile fracture with unstable borehole wall and fault activation, shear fracture with unstable borehole wall and fault activation, and tensile-shear fracture with unstable borehole wall and fault activation; the judgment method is that: based on the observation results of the imaging means, the fault at the natural or artificial weak plane does not shear or open, which is "stable borehole wall without fault activation"; the fault at the natural or artificial weak plane shears, which is "shear fracture with unstable borehole wall and fault activation"; the fault at the natural or artificial weak plane opens, which is "tensile fracture with unstable borehole wall and fault activation"; and the fault at the natural or artificial weak plane opens and shears, which is "tensile-shear fracture with unstable borehole wall and fault activation";

[0077] S10, repeating the experiment to obtain a large amount of data, performing principal component analysis according to the data of the experiment to obtain weight coefficients, and then establishing a relationship model between the borehole orientation, natural weak plane, stress state, temperature, wellbore pressure and fault activation / natural weak plane activation state and the borehole stability in the central control module 6, analyzing the influence mechanism of fault activation on borehole stability and the control method of avoiding fault activation / natural weak plane activation under drilling conditions.

[0078] Specifically, in S10, a group of data is recorded at the natural fracture on each borehole wall in each experiment, which is a row in the data table, wherein the input parameters of the group of data are the borehole orientation at the intersection of the natural fracture and the borehole wall, the angle between the natural weak plane and the wellbore, the stress state, the temperature, and the wellbore pressure, and the output parameters are the activation state of the natural weak plane and the micro-fracture, and the borehole stability; different natural fractures on different boreholes in different experiments are counted as different rows in the data set;

[0079] The borehole orientation and the angle between the natural weak plane and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak plane; the stress state, the temperature, and the wellbore pressure are obtained through the sensors arranged inside the test core 1;

[0080] The data set is obtained by repeatedly changing the parameters and repeating the experiment;

[0081] The data set is sorted, and the well wall orientation, the angle between the natural weak plane and the wellbore, the stress state, the temperature, and the wellbore pressure parameter are normalized to 0-1 by using a normalization method, and the activation state of the natural weak plane and the micro crack, and the well wall stability are classified according to the previous step, and the well wall is stable and no fault activation is attached with a weight value of 0; the well wall is unstable and the fault activation shear joint is attached with a weight value of 1; the well wall is unstable and the fault activation opening joint is attached with a weight value of 2; the well wall is unstable and the fault activation tensile-shear joint is attached with a weight value of 3; and a sorted data set is obtained;

[0082] The weight of the input parameter is analyzed by using the principal component analysis method, and the weight coefficient is obtained, which is sorted in descending order, and the greater the weight, the more significant the influence of the parameter on the fault activation and the natural weak plane activation.

[0083] Specifically, in S10,

[0084] A machine learning classification model is established to realize prediction of the activation state of the natural weak plane and the micro crack and the well wall stability.

[0085] In the model construction process, a group of data is recorded at the place with a natural crack on each well wall of each experiment, which is taken as a row in the data table, wherein the input parameters of the group of data are the well wall orientation at the intersection of the natural crack and the well wall, the angle between the natural weak plane and the wellbore, the stress state, the temperature, and the wellbore pressure, and the output parameters are the activation state of the natural weak plane and the micro crack and the well wall stability; different natural cracks on different well walls of different experiments are counted as different rows in the data set;

[0086] The well wall orientation and the angle between the natural weak plane and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak plane; the stress state, the temperature, and the wellbore pressure are obtained through the sensors arranged inside the test core 1;

[0087] The data set is obtained by repeatedly changing the parameters and repeating the experiments;

[0088] In the process of establishing the machine learning classification model, the data set is used as the training model, that is, the input parameters are the well wall orientation at the intersection of the natural crack and the well wall, the angle between the natural weak plane and the wellbore, the stress state, the temperature, and the wellbore pressure, and are normalized, the output parameters are the activation state of the natural weak plane and the micro crack and the well wall stability, and are marked by assigning weights in four categories; the input parameters are used as the sample data set, and the corresponding output parameters are used as the label data set, and machine learning training is performed to obtain the trained machine learning classification model.

[0089] When the machine learning classification model is applied, a real-time evaluation of the wellbore stability under the fault activation condition is realized, that is, based on the field logging interpretation data, on the target layer with the same rock as the experiment, the intersection of the natural fracture and the wellbore orientation, the angle between the natural weak plane and the wellbore, the stress state, the temperature, the wellbore pressure and other parameters on the logging interpretation curve are normalized in the same way as in the experiment, and are used as input parameters, and the activation state of the natural weak plane and the micro fracture and the prediction weight of the wellbore stability are obtained through the trained machine learning classification model, so that the prediction of whether the natural weak plane and the micro fracture are activated and whether the wellbore is stable in the real wellbore is realized.

[0090] The fault activation mechanism is analyzed through the prediction model.

[0091] In the specific embodiment, in S2 and S9, the imaging means include a camera, a small wellbore imaging system, CT and nuclear magnetic resonance.

[0092] The design principle of the present application is as follows:

[0093] In the indoor construction of the sample containing the natural weak plane and the micro fracture, the simulated wellbore 2 of different directions and different shapes is manufactured, different confining pressures and temperatures are applied on the test core 1, and the drilling fluid pressure is applied inside the simulated wellbore 2, so as to reproduce the environment encountered by the wellbore in the drilling process of the shallow shale, and then the activation of the micro fracture and the natural weak plane in each simulated wellbore 2 is counted, and the stability of the simulated wellbore 2 after the fault activation / natural weak plane activation is counted, a large amount of data is obtained through a large number of repeated experiments, a database of the fault activation / natural weak plane activation condition under the conditions of the formation temperature, stress and fracturing fluid soaking state is constructed, and a database of whether the wellbore is stable under the condition of the fault activation / natural weak plane activation is constructed, which is used for further analyzing the influence mechanism of the fault activation on the wellbore stability. Finally, the present application combines with the numerical simulation of the field drilling to evaluate the risk of the fault activation / natural weak plane activation to the drilling process, and through the real-time feedback data of the field drilling, the risk of the wellbore stability failure caused by the fault activation / natural weak plane activation is predicted, and the control method of avoiding the fault activation / natural weak plane activation under the drilling condition is analyzed. In addition, the drilling parameters can be modified and optimized in the drilling design process to achieve the purpose of controlling or avoiding the negative influence of the fault activation on the wellbore stability.

[0094] The components not described in detail in the present embodiment are existing components that can be purchased in the public channel.

[0095] The above description of the embodiments is to enable those of ordinary skill in the art to understand and use the invention. Those skilled in the art can easily make various modifications to the embodiments and apply the general principles described herein to other embodiments without creative effort, which should be within the scope of the invention. Therefore, the invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the invention without departing from the scope of the invention should be within the scope of the invention.

[0096] According to the above method, the wellbore stability under fault activation conditions is predicted for 10 natural fractures of X well using the trained machine learning classification model, and the predicted data is verified using the core observation method. The verification results are as follows: the statistical accuracy rate reaches 90%. Therefore, the quantitative evaluation and prediction of wellbore stability under fault activation conditions can be realized by the method.

[0097] Serial number Predicted weight Actual weight Interpretation Whether accurate 1 0 0 Wellbore stable and no fault activation Yes 2 2 2 Wellbore unstable and open fracture of fault activation Yes 3 2 2 Wellbore unstable and open fracture of fault activation Yes 4 2 2 Wellbore unstable and open fracture of fault activation Yes 5 3 3 Wellbore unstable and tensile-shear fracture of fault activation Yes 6 0 0 Wellbore stable and no fault activation Yes 7 3 3 Wellbore unstable and tensile-shear fracture of fault activation Yes 8 2 3 Wellbore unstable and tensile-shear fracture of fault activation No 9 3 3 Wellbore unstable and tensile-shear fracture of fault activation Yes 10 0 0 Wellbore stable and no fault activation Yes

Claims

1. A device for simulating and evaluating wellbore stability under fault activation conditions, characterized in that, include: Test core (1), simulated well wall (2), simulated wellhead (3), confining pressure application and temperature control application module (4), manifold control module (5), central control module (6); The test core (1) has a natural or artificial weak surface inside; The simulated well wall (2) is a simulated well wall drilled inward from the wall of the test core (1) to simulate the open hole wall of an uncemented well. The opening of the simulated well wall (2) includes at least three mutually perpendicular surfaces to simulate different drilling orientations downhole. Each simulated well wall (2) on each surface includes at least two wells that are perpendicular to and inclined to the wall of the test core (1). Each simulated well wall (2) intersects with a fault. The simulated wellhead (3) is installed at the opening of the simulated well wall (2) on the wall of the test core (1). One end of the simulated wellhead (3) is connected to the manifold control module (5) through a pressure manifold. The manifold control module (5) can inject high-pressure fluid into each simulated well wall (2) and maintain the pressure. The confining pressure application and temperature control application module (4) is located outside the test core (1). The confining pressure application and temperature control application module (4) can apply confining pressure to the outer wall of the test core (1) and can heat the test core (1) and keep the temperature of the test core (1) constant. The central control module (6) is connected to the confining pressure application and temperature control application module (4) and the manifold control module (5) respectively. The central control module (6) controls the confining pressure application and temperature control application module (4) and the manifold control module (5) to change the confining pressure, temperature and fluid pressure in the simulated well wall (2), and judge the fault activation status in each direction and its impact on well wall stability. This is used to guide the analysis of the impact mechanism and control method of fault activation on well wall stability.

2. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The test core (1) is a natural outcrop or an artificial core.

3. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The simulated well walls (2) described above cannot be interconnected or directly connected to the test core (1).

4. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The simulated wellhead (3) is sealed to the simulated well wall (2) and the pressure manifold respectively.

5. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The confining pressure application and temperature control application module (4) can apply true triaxial pressure to the outer wall surface of the test core (1).

6. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The central control module (6) can control the manifold control module (5) to apply liquid pressure to each simulated well wall (2) in the order, magnitude and duration.

7. The wellbore stability simulation and evaluation device under fault activation conditions according to claim 1, characterized in that, The central control module (6) can simultaneously control the magnitude, sequence, and time of the confining pressure applied to the test core by the confining pressure application module (4) and the temperature control application module (4); The central control module (6) can simultaneously control the temperature and constant temperature control applied to the test core by the confining pressure application and temperature control application module (4).

8. A method for simulating and evaluating wellbore stability under fault activation conditions, characterized in that, Includes the following steps: S1. Select an outcrop of the target reservoir rock sample or manufacture an artificial core, and process it into a square test core (1) for later use. The test core (1) includes a natural weak surface or an artificial weak surface. S2. Drill holes on each face of the test core (1) to create a simulated well wall (2). The opening of the simulated well wall (2) includes at least three mutually perpendicular faces of the simulated well wall (2) to simulate different drilling positions downhole. Each simulated well wall (2) on each face includes at least two wells that are perpendicular to and inclined to the wall of the test core (1). Each simulated well wall (2) intersects with a fault. The micro-crack state inside each simulated well wall (2) is photographed and archived by imaging means. S3. Install a simulated wellhead (3) on each simulated well wall (2) and connect it to the manifold control module (5) through a pressure manifold; S4. Place the test core (1) inside the confining pressure application and temperature control application module (4); S5. The confining pressure application and temperature control application module (4) are controlled by the central control module (6) to heat the test core (1) so that it reaches the set constant temperature state to simulate the formation conditions. S6. Write the instructions for the magnitude, sequence and duration of the confining pressure and fracturing fluid application for each simulated wellhead (3) in the central control module (6); S7. Start the experiment. The confining pressure application and temperature control application module (4) applies confining pressure to the test core to simulate the formation stress state. The manifold control module (5) injects liquid into each simulated well wall (2) and applies pressure to simulate the open hole well wall during the drilling process. After the instructions in the central control module (6) are applied, the confining pressure and the fluid pressure in the wellbore are slowly removed to prevent stress change from causing secondary damage to the simulated wellbore (2) and the test core (1) is taken out. S9. Using imaging methods, the microcrack state inside each simulated well wall (2) is photographed and archived again. The activation state of the natural weak surface and microcrack is determined by the displacement of the microcracks on the simulated well wall (2). The stability of the simulated well wall (2) is determined and recorded by the fracture, defects and falling blocks of the simulated well wall (2). The damage state of the microcracks on the simulated well wall (2) is divided into four categories: stable well wall and no fault activation, unstable well wall and fault activation tensile crack, unstable well wall and fault activation shear crack, and unstable well wall and fault activation tensile-shear crack. S10. Repeat the experiment to obtain a large amount of data. Perform principal component analysis based on the experimental data to obtain the weight coefficients. Analyze the influencing factors and mechanisms of wellbore stability through the weight coefficients. Then, establish a model in the central control module (6) to establish the relationship between wellbore orientation, natural weak surface, stress state, temperature, wellbore pressure and fault activation / natural weak surface activation state and wellbore stability. Further analyze the influence mechanism of fault activation on wellbore stability and the control method to avoid fault activation / natural weak surface activation under drilling conditions.

9. The wellbore stability simulation and evaluation method under fault activation conditions according to claim 8, characterized in that, In S10, a set of data is recorded for each location with natural cracks on the wellbore in each experiment. This data is recorded as a row in the data table. The input parameters for this set of data are the wellbore orientation at the intersection of the natural crack and the wellbore, the angle between the natural weak surface and the wellbore, the stress state, the temperature, and the wellbore pressure. The output parameters are the activation state of the natural weak surface and the microcracks, and the wellbore stability. Different natural cracks on different wellbores in different experiments are counted as different rows in the dataset. The wellbore orientation and the angle between the natural weak surface and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak surface; the stress state, temperature, and wellbore pressure are obtained through sensors arranged inside the test core (1); By repeatedly experimenting with constantly changing parameters, a dataset can be obtained. The dataset is processed by normalizing the wellbore orientation, the angle between the natural weak surface and the wellbore, the stress state, temperature, and wellbore pressure parameters to between 0 and 1. The activation state of the natural weak surface and microfractures, and the wellbore stability are classified according to the previous step: "Wellbore stable and without fault activation" is assigned a weight of 0; "Wellbore unstable and with fault activation (shear fracture)" is assigned a weight of 1; "Wellbore unstable and with fault activation (open fracture)" is assigned a weight of 2; and "Wellbore unstable and with fault activation (tensile-shear fracture)" is assigned a weight of 3. This yields the processed dataset. Principal component analysis was used to analyze the weights of the input parameters and obtain the weight coefficients. The coefficients were sorted from largest to smallest. The larger the weight, the more significant the influence of the parameter on fault activation and natural weak surface activation. The weighting coefficient is a quantitative characterization of wellbore stability under fault activation conditions. The influencing factors and mechanisms of wellbore stability are analyzed through the weighting coefficient.

10. The wellbore stability simulation and evaluation method under fault activation conditions according to claim 8, characterized in that, In S10, Establish a machine learning classification model to predict the activation state of natural weak surfaces and microfractures, as well as wellbore stability; During model building, a set of data was recorded for each location with natural cracks on the wellbore in each experiment, and this data was recorded as a row in the data table. The input parameters for this set of data were the wellbore orientation at the intersection of the natural crack and the wellbore, the angle between the natural weak surface and the wellbore, the stress state, the temperature, and the wellbore pressure. The output parameters were the activation state of the natural weak surface and the microcracks, and the wellbore stability. Different natural cracks on different wellbores in different experiments were counted as different rows in the dataset. The wellbore orientation and the angle between the natural weak surface and the wellbore are obtained through the geometric relationship between the wellbore and the natural weak surface; the stress state, temperature, and wellbore pressure are obtained through sensors arranged inside the test core (1); By repeatedly experimenting with constantly changing parameters, a dataset can be obtained. In the process of establishing the machine learning classification model, the dataset is used as the training model. The input parameters are the well wall orientation at the intersection of natural fractures and well wall, the angle between the natural weak surface and the wellbore, stress state, temperature, and wellbore pressure. After normalization, the output parameters are the activation state of the natural weak surface and micro fractures, and well wall stability. They are divided into four categories and labeled with weights. The input parameters are used as the sample dataset, and the corresponding output parameters are used as the label dataset. Machine learning training is performed to obtain the trained machine learning classification model. When the above-mentioned machine learning classification model is applied, a real-time evaluation of wellbore stability under fault activation conditions is achieved. That is, based on field logging interpretation data, on the target layer with the same lithology as the experiment, the trained machine learning classification model is used to normalize parameters such as the wellbore orientation at the intersection of natural fractures and wellbore on the logging interpretation curve, the angle between the natural weak surface and the wellbore, stress state, temperature, and wellbore pressure, in the same way as in the experiment, and use them as input parameters. The trained machine learning classification model obtains the corresponding activation state of natural weak surfaces and microfractures, and the predicted weighted values ​​of wellbore stability, and finally realizes the prediction of whether natural weak surfaces and microfractures are activated and whether the wellbore is stable in the real wellbore. The mechanism of fault activation was analyzed using a predictive model.

Images