Earthquake prediction-based drilling collapse jamming early warning method

Abstract

The invention belongs to the technical field of oil exploration, and particularly relates to a well drilling collapse and jamming early warning method based on earthquake prediction, which comprises the following steps: firstly, determining easily collapsed lithology, and then identifying an easily collapsed lithology layer according to a post-stack earthquake data body and conventional logging information; determining a fault or crack or fault crack development area of the easily collapsed lithologic layer; and fusing the designed well trajectory with the obtained fault or crack or fault crack development area of the easy-to-collapse lithologic layer, determining the overlapped part of the well trajectory and the fault or crack or fault crack development area of the easy-to-collapse lithologic layer as a risk well section, and performing early warning on the position of the risk well section. As the best method for reducing the drilling tool jamming loss is to give an early warning for the drilling tool jamming accident in advance, the prediction process is carried out when the well trajectory is designed, so that the situation of drilling collapse and drilling tool jamming at which position exists is predicted after the well trajectory is designed and before drilling, the early warning before drilling is realized, and the accuracy of drilling tool jamming is improved. The early warning is high in timeliness.

Description

Technical Field

[0001] The invention belongs to the technical field of petroleum exploration, and in particular relates to a drilling collapse and pipe sticking early warning method based on earthquake prediction. Background Art

[0002] The oil drilling process is a complex project with a lot of randomness, ambiguity and uncertainty. During the drilling process, the phenomenon that the drill bit is stuck in the well and cannot move freely due to various reasons is called stuck drill. Drill stuck is a common underground accident, and the accident handling process is relatively complicated. Once the drill bit is stuck, it will cause the drilling cycle to be extended and the drilling cost to be increased. Improper handling may also cause other underground accidents such as collapse and drill bit falling, and even cause the wellbore to be scrapped, resulting in huge economic losses. According to the article "Handling and Analysis of Complex Drill Bits Stuck in AD401-7 Directional Well" by Wei Na (2018 "Exploration Engineering (Rock and Soil Drilling Engineering)" Vol. 45 No. 4), and the article "Mechanism and Countermeasures of Shale Wellbore Instability in Indonesia M Gas Field" by Jiang Kai (2021 "China Offshore Oil and Gas" Vol. 33 No. 2), under normal operating conditions, drilling in broken strata such as thin coal seam interlayers or mudstone micro-faults and fractures is most likely to cause collapse and stuck drill. In addition to the thin coal seam interlayer or mudstone mentioned in the article, rock salt and uncemented rocks may also collapse and get stuck. Therefore, accurate early warning of the location of the stuck drill in oil drilling can provide basic data for wellbore structure design, determination of drilling fluid density, protection of oil and gas layers, and improvement of drilling success rate.

[0003] At present, with the extension of mining time, the reality of decreasing underground reserves is inevitable. The local structure near the fault has become a favorable area for fine tapping of remaining oil and gas, and the well layout has changed from "avoiding faults" to "finding faults-relying on faults-penetrating faults". Affected by faults and fractures controlled by faults, drilling near faults is more risky and more prone to collapse and stuck pipe accidents. At present, for possible collapse and stuck pipe accidents during drilling, the main method used is the detection and early warning method of stuck pipe accidents based on logging big data. This method mainly involves engineering anomaly detection, which has poor timeliness and increases the probability of risk occurrence and the cost of handling risks. In addition, the near-fault area is a small well area with low drilling density. Due to insufficient logging data, there is also a lack of learning samples, which reduces the accuracy of detection and early warning of stuck pipe accidents based on logging big data.

[0004] In addition, in order to provide early warning for stuck drill, oil fields often use integrated logging instruments to collect real-time downhole data, manually set the thresholds of parameters related to stuck drill, and issue a stuck drill alarm when the parameters collected in real time exceed the set thresholds. The diagnosis of the type of stuck drill still relies on the work experience of logging data analysts to make a preliminary judgment. However, this early warning method is still almost the same as the time of the accident, and it does not achieve the purpose of avoiding serious stuck drill accidents and improving drilling efficiency through early warning for stuck drill. Moreover, the diagnosis of the type of stuck drill has certain human factors, and may cause misjudgment due to experience limitations.

[0005] For example, the Chinese invention patent application with application publication number CN110778307A provides a method for detecting and warning of drill stuck accidents based on logging big data, which involves engineering anomaly detection, including preprocessing the logging data and the log data of the management personnel to realize automatic labeling of drilling accidents, and resampling and smoothing the original data, and then processing the drill stuck data according to the random forest model, and finally adjusting the learning model parameters using the cross-validation method to generate a drill stuck detection model. In the case of less developed stable faults in the formation, this patent can achieve a relatively accurate warning of drill stuck accidents, but microfaults in the near-fault area are the main cause of collapse, and logging data is difficult to accurately locate faults. In addition, the location of the adjacent drilling fault is not representative for the next well. In the absence of accurate positioning of the microfaults, the decision tree of the random forest model is difficult to construct, and it is difficult to accurately warn of the drill stuck location. Although conventional seismic prediction methods can predict microfaults and cracks in well deployment work, such as "The concept of "fault body", geological model and its application in crack prediction-taking the second segment of the tight sandstone gas reservoir of the Xujiahe Formation in Xinchang area, western Sichuan Depression, Sichuan Basin as an example" by Liu Zhenfeng (Volume 42, No. 4 of "Petroleum & Natural Gas Geology" in 2021), it is mentioned that the corresponding seismic attribute thresholds are determined by using the interpretation results of artificial faults and electrical imaging fractures in oil and gas exploration in Xinchang area, and the maximum likelihood attribute (>0.15) is used to characterize the fault sliding surface, and the different threshold ranges of chaotic attributes are used to characterize the sliding fracture zone (0.18~1.00) and the induced fracture zone (0.05~0.18) to predict the distribution range of the fault sliding surface, sliding fracture zone and induced fracture zone. However, since the collapse-prone layer itself is not a single lithology, and the brittleness of the collapse-prone lithology layer is quite different from that of the surrounding rock, if the fractures are calibrated according to regional parameters, the prediction results will cause misjudgment. Summary of the invention

[0006] The purpose of the present invention is to provide a drilling collapse and pipe stuck early warning method based on earthquake prediction, so as to solve the problem of poor timeliness of the existing pipe stuck early warning methods.

[0007] In order to solve the above technical problems, the present invention provides a drilling collapse and stuck pipe early warning method based on earthquake prediction, comprising the following steps:

[0008] 1) Obtain geological and seismic data of the target work area to determine the lithology prone to collapse, and identify the lithology layer prone to collapse based on the post-stack seismic data and conventional logging data in the geological and seismic data;

[0009] 2) Determine the faults or cracks or fault-crack development areas of the collapse-prone lithology layer based on the post-stack seismic data volume;

[0010] 3) In the drilling geological design, the designed well trajectory is integrated with the obtained faults or cracks or fault-crack development areas of the easily collapsed lithology layer, and the part where the well trajectory overlaps with the faults or cracks or fault-crack development areas of the easily collapsed lithology layer is determined as a risky well section, and an early warning is issued for the location of the risky well section.

[0011] The beneficial effects are as follows: the method of the present invention takes into account that the best way to reduce stuck drill losses is to give early warning of stuck drill accidents. Therefore, the method of the present invention performs a prediction process when designing the well trajectory, so that after the well trajectory is designed and before drilling, it is predicted at which position the drilling collapse and stuck drill situation may occur, thereby achieving early warning before drilling. The warning has strong timeliness, that is, it can be warned before the failure occurs. Compared with the prior art in which the warning time and the drilling collapse and stuck drill situation can only be identified during drilling, this method has strong timeliness in warning.

[0012] Furthermore, in step 1), the method for identifying the collapse-prone lithology layer is: determining the recognition threshold of collapse-prone lithology through conventional logging data, performing wave impedance inversion using post-stack seismic data, and using the wave impedance inversion data to identify the collapse-prone lithology layer within the threshold.

[0013] The method of the present invention obtains the identification threshold method first, that is, by obtaining the identification threshold corresponding to different easily collapsible rock properties, and then based on this identification threshold, when judging, if it falls within the range of the identification threshold, it is the easily collapsible rock property layer corresponding to the identification threshold.

[0014] Furthermore, the method for determining the identification threshold is: obtaining a wave impedance curve by calculating conventional logging data, determining the wave impedance range of the collapse-prone lithology according to the wave impedance curve, and using the wave impedance range as the identification threshold.

[0015] Since the wave impedance range is different in different lithologies, different lithologies can be judged by wave impedance. Therefore, in the present invention, the wave impedance range of the collapse-prone lithology is determined by obtaining a wave impedance curve, and then the wave impedance range is used as an identification threshold to identify the lithology corresponding to the wave impedance range.

[0016] Furthermore, in step 2), the method for determining the fault development zone of the prone-to-collapse well section is as follows: extracting coherent attributes in the post-stack seismic data volume, comparing conventional logging data with breakpoint data, and calibrating the coherent attributes using the breakpoint data to obtain the fault development boundary thresholds of the fault development positions and non-fault development positions of each prone-to-collapse well section, and determining the fault development zone of the prone-to-collapse well section based on the fault development boundary thresholds.

[0017] Furthermore, in step 2), the method for determining the fracture development zone of the collapse-prone well section is as follows: extracting curvature attributes in the post-stack seismic data volume, obtaining the fracture development section according to the fracture interpretation results of the well imaging logging interpretation in the seismic geological data, calibrating the curvature attributes using the obtained fracture development section, determining the boundary thresholds of fracture development of different collapse-prone lithologies, and delineating the fracture development zone by the curvature attributes and the boundary thresholds of fracture development of different collapse-prone lithologies.

[0018] In the present invention, coherent attributes are extracted from post-stack seismic data to predict the location of fault development, and curvature attributes are further extracted to predict the fracture development zone formed by folds. Seismic attribute thresholds are determined according to different lithologies, and the location of fault and fracture development in each easily collapsed lithology is identified to obtain the location of the fault or fracture or fault-fracture development zone in the easily collapsed lithology layer, which is prone to drilling collapse and drill jamming.

[0019] Furthermore, in step 3), the easily collapsed lithology layer is first merged with the faults or cracks or the fault-crack development zone of the easily collapsed lithology layer to obtain a prediction result data body, and then the designed well trajectory is added to the prediction result data body to obtain the part where the well trajectory overlaps with the faults or cracks or the fault-crack development zone of the easily collapsed lithology layer.

[0020] In the method of the present invention, the final collapse-prone layer section prediction result is obtained through data fusion. The fusion process can intuitively and accurately display the location of the risky well section, thereby making advance predictions during drilling geological design.

[0021] Furthermore, the designed well trajectory includes designed coordinates, designed well deviation, designed layering and time-depth relationship.

[0022] Furthermore, in step 4), when the well trajectory overlaps with the collapse-prone lithology layer fault or crack or the fault-crack development zone, the corresponding time depth is obtained, and an early warning is issued for the risk well section position according to the time depth.

[0023] Furthermore, by setting a time value before the time depth, an early warning is issued for the risk section location.

[0024] The method of the present invention uses the time-depth relationship as the designed well trajectory, thereby accurately obtaining the time depth at which the risk well section will be drilled, and then implementing an effective and accurate early warning process by issuing an early warning before the time.

[0025] Furthermore, in step 1), the easily collapsible lithology includes at least one of mudstone and coal seam strata. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of the drilling collapse and stuck pipe early warning method based on earthquake prediction of the present invention;

[0027] Figure 2 is a wave impedance inversion cross-section diagram in this embodiment;

[0028] Figure 3 is a cross-sectional view of the coherent properties of Well A in this embodiment;

[0029] Figure 4 This is a cross-sectional diagram of the prediction results of the mudstone and coal seam fault fractures in Well A of this embodiment;

[0030] Figure 5 This is a cross-sectional diagram of the prediction results of the mudstone and coal seam fault fractures in Well B of this embodiment. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0032] Embodiment of drilling collapse and stuck pipe early warning method based on earthquake prediction:

[0033] This embodiment takes the Qianfoya Formation and Xuwu Section exploration area of ​​the MB fault zone in northeastern Sichuan as an example. Faults are well developed in this area, the existing drilling density is not high, and the basic conditions for similar logging big data prediction are lacking. The method provided by the present invention is used to illustrate the method for early warning of stuck pipe.

[0034] Depend on Figure 1 The specific steps are as follows:

[0035] 1) Collect geological and seismic data of the target work area to determine the lithology of the well section prone to collapse.

[0036] The geological and seismic data include: drilling construction records, logging data, and well logging data of the wells drilled in the target work area and nearby areas, including natural gamma curves, sonic logging data, density logging data, fracture interpretation results of imaging logging interpretation of the wells that have been drilled, and post-stack seismic data volumes; the lithology of the well sections prone to collapse in the study area is determined by drilling construction records and logging data, and they are collectively defined as L1, L2, ... Ln, where n is the total number of lithologies of well sections prone to collapse.

[0037] In this embodiment, basic data of the study area are collected and counted. The well data include logging data (acoustic time difference curve, density curve, lithology interpretation results, resistivity curve, fracture interpretation results of imaging logging interpretation) of a well (Well A) drilled in the vicinity of the fault structure belt and data of a target well (Well B) (design coordinates, design well inclination, design stratification, time-depth relationship); seismic data include post-stack seismic data volume; regional geological interpretation results include stratigraphic and breakpoint data, and mudstone development intervals.

[0038] The main collapse-prone strata in the study area are the mudstone of the Qianfoya Formation and the coal seam of the Xuwu Member.

[0039] 2) Use post-stack seismic data and conventional logging data to identify mudstone and coal seams.

[0040] 2.1) Rock physical analysis was carried out using conventional logging data, and the thresholds for identifying different combinations of collapse-prone lithologies were established, which were defined as A1, A2, …, An.

[0041] The wave impedance curve is calculated by using the acoustic time difference and density curve. The formula is Pimp = 1000000 / Ac*den (where Ac is the logging acoustic time difference, unit is us / m, and den is the density, unit is g / cm 3 , Pimp is wave impedance, unit is g / cm 3 *m / s). As another implementation method, the wave impedance curve can be calculated by conventional logging data, and then the wave impedance curve histogram can be obtained to establish mudstone and coal seam identification thresholds, and then the wave impedance data volume can be obtained by the inversion method, and the mudstone and coal seam identification thresholds can be used to identify mudstone and coal seams.

[0042] 2.2) Calculate the wave impedance curve using the acoustic logging data and density logging data, and use the wave impedance curve to establish a wave impedance low-frequency model, use the post-stack seismic data volume to perform wave impedance inversion, and use the threshold established in step 2.1) to predict the collapse-prone lithology layer.

[0043] Specifically, in this embodiment, the impedance curve calculated in step 2.1) is applied, combined with the lithology interpretation results collected in step 1), to establish the impedance threshold for mudstone identification as 9700-11200 g / cm 3 *m / s, defined as a1, the wave impedance threshold for coal seam identification is less than 9700g / cm 3 *m / s, defined as a2.

[0044] The wave impedance curve calculated in step 2.1) is used to establish a low-frequency wave impedance model. The wave impedance inversion is performed using the post-stack seismic data collected in step 1) to obtain a wave impedance data volume. The established wave impedance threshold a1 is used to predict mudstone, and the wave impedance threshold a2 is used to predict coal seams. Figure 2Medium wave impedance 9700-11200g / cm 3 *m / s is the mudstone layer, and the wave impedance is less than 9700g / cm 3 *m / s refers to the coal seam development interval, such as Figure 2 The Well01 in the figure is a schematic diagram of Well A.

[0045] 3) Predict fault fracture development areas.

[0046] 3.1) Apply the post-stack seismic data collected in step 1) to extract coherent attributes, compare the breakpoint data with the conventional logging data collected in step 1), and use the breakpoint data to calibrate the coherent attributes to determine the development locations of different collapse-prone lithology faults and non-fault development locations (the coherent threshold attribute values ​​of the development boundaries of different lithology faults are defined as B1, B2, ..., Bn).

[0047] In this embodiment, the post-stack seismic data collected in step 1) is used to extract coherent attributes, and the breakpoint data is compared with the conventional logging data collected in step 1), and the coherent attributes are calibrated using the breakpoint data to determine the coherent attribute threshold of 0.86 when the coal seam fault is developed. The value less than 0.86 indicates the fault development position, and the fault development boundary threshold attribute value is defined as b1 (such as Figure 3 ).

[0048] 3.2) Apply the post-stack seismic data collected in step 1) to extract curvature attributes, obtain the fracture development section according to the fracture interpretation results of the well imaging logging interpretation collected in step 1), calibrate the curvature attributes using the well fracture development section, determine the fracture development boundary thresholds of different collapse-prone lithologies (the curvature attribute values ​​of the fracture development boundaries are defined as C1, C2, ..., Cn), and delineate the fracture development area using the curvature attributes and the fracture development boundary thresholds of different collapse-prone lithologies.

[0049] In this embodiment, the curvature attribute is extracted from the post-stack seismic data volume collected in step 1), and the curvature attribute is calibrated according to the fracture interpretation results of the imaging logging interpretation collected in step 1), and the fracture interpretation results of the imaging logging interpretation are used to determine that when mudstone fractures develop, the curvature attribute threshold is 0.00092, and a value greater than 0.00092 indicates a fracture development location (the fracture development boundary threshold attribute value is defined as c1), and when coal seam fractures develop, the curvature attribute threshold is 0.00052, and a value greater than 0.00052 indicates a fracture development location, and the fracture development boundary threshold attribute value is defined as c2.

[0050] 4) The predicted results of the fault development position predicted in step 3) are merged with the fracture development area to obtain the fault fracture prediction result data body of the study area, and then the collapse-prone rock layer predicted in step 2) is merged with the fault fracture prediction result data body to obtain the collapse-prone rock layer fault fracture prediction result data body.

[0051] In this embodiment, the mudstone and coal seam strata calculated in step 2) and the fault and fracture development zone in step 3) are merged to obtain the fault and fracture prediction result data body of the mudstone and coal seam near well A (such as Figure 4 ), mudstone collapse occurred in Well A when drilling to 1449-1463ms, and the drill got stuck when drilling to the coal seam at 1729-1745ms, which is the same as the result obtained by the early warning method in this embodiment.

[0052] 5) During the drilling geological design, the designed well trajectory is added to the collapse-prone lithology fault fracture prediction result data body obtained in step 4 to determine the potential risk well section and to warn of the potential collapse and stuck drill risk location.

[0053] Steps 2), 3), and 4) are used to obtain the fault and fracture prediction result data body of the mudstone and coal seam near the target well B in the area with few wells near the fault. In the drilling geological design, the design coordinates, design well inclination, design layering, and time-depth relationship of the B well collected in step 1) are added to the fault and fracture prediction result data body of the mudstone and coal seam near the B well, and the potential risk well sections with mudstone and coal seam formations at the time depth of 740-753ms and 1033-1049ms and developed faults or fractures or both faults and fractures are determined, so as to warn in advance the risk of drilling stuck in the fracture development section of drilling, such as Figure 5 The Well02 in the figure is a schematic diagram of Well B.

[0054] The method in this embodiment takes into account that the best way to reduce the loss of stuck drill is to give early warning of the stuck drill accident to provide a decision-making basis for the unblocking operation. Therefore, the method in this embodiment is aimed at the formation fracture zone formed near the local or regional faults and folds, and the situation of collapse and stuck drill in unstable strata. The collapse-prone lithology is determined by regional geological data, and the collapse-prone lithology layer is identified by lithology using post-stack seismic data and conventional logging data. The coherent attributes are extracted from the post-stack seismic data to predict the development position of the fault, and the curvature attributes are further extracted to predict the fracture development zone formed by the fold. The seismic attribute threshold is determined according to different lithologies, and the fault and fracture development position of each collapse-prone lithology is judged. Finally, the final collapse-prone layer prediction result is provided through data fusion, and the prediction is made in advance during the drilling geological design. This method provides an early warning of collapse and stuck drill during geological design, providing an important reference for the wellbore structure design and the determination of drilling fluid density. Compared with the detection and early warning method of stuck drill accidents based on logging big data, it is more targeted and timely, and can reduce the probability of risk occurrence and the time cost of dealing with risks.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. The patent protection scope of the present invention shall be based on the claims. All equivalent structural changes made using the contents of the description and drawings of the present invention should also be included in the protection scope of the present invention.

Claims

1. A drilling collapse and stuck pipe early warning method based on earthquake prediction, characterized in that: The steps include: 1) Obtain geological and seismic data of the target work area to determine the lithology prone to collapse, and identify the lithology layer prone to collapse based on the post-stack seismic data and conventional logging data in the geological and seismic data; 2) Determine the faults or cracks or fault-crack development areas of the collapse-prone lithology layer based on the post-stack seismic data volume; 3) In the drilling geological design, the designed well trajectory is integrated with the obtained faults or cracks or fault-crack development areas of the easily collapsed lithology layer, and the part where the well trajectory overlaps with the faults or cracks or fault-crack development areas of the easily collapsed lithology layer is determined as a risky well section, and an early warning is issued for the location of the risky well section.

2. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 1, characterized in that: In step 1), the method for identifying the collapse-prone lithology layer is: determining the recognition threshold of collapse-prone lithology through conventional logging data, performing wave impedance inversion using post-stack seismic data, and using the wave impedance inversion data to identify the collapse-prone lithology layer within the threshold.

3. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 2, characterized in that: The method for determining the identification threshold is: obtaining a wave impedance curve by calculating conventional logging data, determining the wave impedance range of the collapse-prone lithology according to the wave impedance curve, and using the wave impedance range as the identification threshold.

4. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 1, characterized in that: In step 2), the method for determining the fault development zone of the prone-to-collapse well section is as follows: extracting coherent attributes in the post-stack seismic data volume, comparing conventional logging data with breakpoint data, and calibrating the coherent attributes using the breakpoint data to obtain the fault development boundary thresholds of the fault development positions and non-fault development positions of each prone-to-collapse well section, and determining the fault development zone of the prone-to-collapse well section based on the fault development boundary thresholds.

5. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 1, characterized in that: In step 2), the method for determining the fracture development zone of the collapse-prone well section is as follows: extracting curvature attributes in the post-stack seismic data volume, obtaining the fracture development section according to the fracture interpretation results of the drilling imaging logging interpretation in the seismic geological data, calibrating the curvature attributes using the obtained fracture development section, determining the boundary thresholds of fracture development of different collapse-prone lithologies, and delineating the fracture development zone by the curvature attributes and the boundary thresholds of fracture development of different collapse-prone lithologies.

6. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 1, characterized in that: In step 3), the prediction result data body is obtained by first merging the easily collapsed lithology layer with the faults or cracks or the area where both faults and cracks are developed in the easily collapsed lithology layer, and then adding the designed well trajectory to the prediction result data body to obtain the part where the well trajectory overlaps with the faults or cracks or the area where both faults and cracks are developed in the easily collapsed lithology layer.

7. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 6, characterized in that: The designed well trajectory includes designed coordinates, designed well deviation, designed layering and time-depth relationship.

8. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 7, characterized in that: In step 4), when the well trajectory overlaps with the collapse-prone lithology layer fault or crack or the fault-crack development zone, the corresponding time depth is obtained, and an early warning is issued for the risk well section position according to the time depth.

9. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 8, characterized in that: By setting the time value before the time depth, an early warning is issued for the risk section location.

10. The drilling collapse and stuck pipe early warning method based on earthquake prediction according to claim 1, characterized in that: In step 1), the easily collapsible lithology includes at least one of mudstone and coal seam strata.

Images