A method and system for predicting the difficulty of reverse-well reverse-pulling construction

Abstract

This invention provides a method and system for predicting the difficulty of reverse well drilling, comprising the following steps: S1, collecting trajectory characteristic parameters during the pilot hole drilling process; S2, preprocessing the collected trajectory data; S3, acquiring the preprocessed trajectory data and extracting trajectory features characterizing the overall spatial morphology and local anomalies of the pilot hole; S4, constructing an evaluation index system reflecting the difficulty of reverse well drilling; S5, establishing an identification and prediction model of the pilot hole trajectory characteristics and the difficulty of reverse well drilling; S6, identifying and predicting the difficulty, risk level, and key challenging sections of reverse well drilling based on the identification and prediction model; S7, outputting the identification and prediction results and generating targeted construction suggestions. This invention can identify potentially high-risk sections and key challenging sections in advance before reverse well drilling, reducing reliance on on-site operational experience, reducing in-process adjustments and post-process handling, and improving the foresight and targeting of construction control.

Description

Technical Field

[0001] This invention relates to the field of reverse well reverse pull construction technology, specifically to a method and system for predicting the difficulty of reverse well reverse pull construction. Background Technology

[0002] Inclined shaft construction typically employs the reverse shaft method, a commonly used well completion technique, especially for inclined shafts. Its basic principle is as follows: first, a pilot hole is formed along the designed axis using a directional drilling rig. Then, an reaming bit is installed at the bottom of the pilot hole, and the hole is enlarged from bottom to top to form a guide hole that meets the requirements of the reverse shaft drilling rig. Finally, the reverse shaft drilling rig is used to excavate the guide shaft. Compared to the traditional drill-and-blast method, the reverse shaft method has advantages such as a high degree of mechanization, better wellbore quality, and higher efficiency, and is now widely used in inclined shaft construction.

[0003] However, in the construction of pilot holes using the reverse drilling method, due to the long construction distance and complex spatial orientation changes, the pilot hole is easily affected by factors such as changes in formation lithology, uneven rock hardness, development of joints and fractures, and distribution of local fracture zones during the drilling process, which can lead to deviations between the actual drilling trajectory and the design axis.

[0004] To understand the drilling status and trajectory changes of the pilot hole, equipment such as inclinometers and measurement-while-drilling systems are often used during construction to obtain the spatial trajectory characteristic parameters of the borehole. These trajectory characteristic parameters are important basic data for evaluating the construction quality of the pilot hole.

[0005] However, currently only some parameters are used to analyze the quality of the pilot hole; and most of the existing construction methods are in-process adjustments or post-event responses after problems occur. There is a lack of technical means to pre-identify and predict the difficulty, risk level and key difficult sections of the subsequent reverse pull construction based on the characteristics of the pilot hole drilling trajectory before the reverse pull construction. There is also a lack of construction parameter optimization, risk warning and construction organization feedforward control mechanism based on this. Summary of the Invention

[0006] The main objective of this invention is to provide a method and system for predicting the difficulty of reverse well construction, thereby solving the problem that existing technologies cannot effectively utilize pilot hole trajectory feature information to make advance judgments on reverse well construction.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: S1. Collect trajectory characteristic parameters during the pilot hole drilling process; S2. Preprocess the collected trajectory data; S3. Obtain preprocessed trajectory data and extract trajectory features that characterize the overall spatial morphology and local anomaly changes of the pilot hole; S4. Construct an evaluation index system that reflects the difficulty of reverse tension construction; S5. Establish an identification and prediction model for the characteristics of the pilot hole trajectory and the difficulty of reverse well reverse pulling construction; S6. Based on the identification and prediction model, identify and predict the difficulty, risk level and key challenging sections of the reverse tension construction. S7. Output the identification and prediction results and generate targeted construction suggestions.

[0008] In the preferred embodiment, in S1: during the drilling of the pilot hole in the inclined shaft, the trajectory characteristic parameters of the pilot hole are collected using a trajectory acquisition device; The trajectory feature parameters include at least 1 to 3 of the following parameters: Well inclination angle I; azimuth angle A; well inclination offset OI; azimuth offset OA; total angle change rate, i.e., dogleg degree D; trajectory curvature C; offset O; Simultaneously collect auxiliary parameters related to the drilling status, including drilling pressure, rotational speed, torque, feed rate, flushing medium flow rate, and flushing medium pressure, for subsequent identification and optimization correction of the prediction model; Trajectory feature parameters and auxiliary parameters are collected at predetermined depth intervals or in real time along the drilling direction of the pilot hole, forming corresponding raw data sequences.

[0009] In the preferred embodiment, in S3: based on the preprocessed trajectory data, trajectory features of the pilot hole are extracted to reflect the overall spatial morphological features and local anomaly change features of the pilot hole. The trajectory features include overall trajectory features, local anomaly features, and comprehensive trajectory features, wherein: Overall trajectory feature extraction is used to characterize the overall orientation and smoothness of the pilot hole, including: the average rate of change of the inclination angle of the entire hole. Average value of the azimuth change rate of the entire borehole ; Overall borehole deviation distribution; Overall borehole azimuth deviation distribution; Overall borehole average dogleg deviation Maximum dogleg width of the full hole Average curvature of the entire hole Total Hole Offset Cumulative Amount ; Local anomaly feature extraction is used to identify local trajectory problems that are detrimental to subsequent reverse pull construction, including: local high dogleg sections; local sections with sudden increases in azimuth offset; local sections with sudden changes in curvature; local sections with sudden changes in eccentricity; continuous length of anomaly sections; and sections with drastic changes in well inclination or azimuth. A section can be identified as an anomaly section if it meets any of the following conditions: ; ; ; ; ; in, , , , , The preset threshold; Let i be the dogleg degree of the i-th measurement point; Let be the curvature of the i-th measurement point; This represents the offset of the i-th measuring point relative to the design axis. Let be the well inclination angle of the i-th measuring point; Let be the azimuth angle of the i-th measuring point; Comprehensive trajectory feature extraction is used to build an identification and prediction model, and further comprehensive trajectory feature indicators are constructed, including: (1) Smoothness index of the entire hole trajectory It is used to comprehensively characterize the continuity, uniformity and stability of the pilot hole trajectory change throughout the entire hole, reflecting whether the entire drilling trajectory is smooth and straight, and whether there are many sharp bends and abrupt changes; the larger the whole hole trajectory smoothness index, the smoother the whole hole trajectory; the smaller the index, the greater the fluctuation of the whole hole trajectory and the more local abrupt changes. (2) Trajectory complexity index It is used to comprehensively reflect the geometric complexity of the pilot hole trajectory throughout the entire hole range, reflecting whether the trajectory changes frequently, whether the spatial shape is complex, and whether the direction adjustment is drastic; the larger the trajectory complexity index, the more complex the trajectory is, and the more likely the stress state of the drill bit is to become more complex during the subsequent reverse pull operation. (3) Local anomaly intensity index It is used to characterize the severity of trajectory anomaly changes within a local segment, focusing on reflecting the concentration of problems such as high dogleg, high curvature abrupt changes, and sudden increases in offset; the local anomaly intensity index is suitable for identifying key and difficult segments and local high-risk segments; (4) Trajectory offset cumulative index It is used to characterize the cumulative degree of overall deviation of the actual trajectory of the pilot hole relative to the design axis, reflecting the cumulative effect of the whole hole deviating from the design axis; the larger the cumulative trajectory deviation index, the more obvious the overall deviation of the trajectory from the design axis, and the greater the potential adverse impact on subsequent reverse tensioning construction. (5) Characteristics of continuous high-risk trajectories It is used to characterize the length and severity of segments in which high-risk trajectory features appear continuously in the pilot hole.

[0010] In the preferred scheme, in S4: an evaluation index system reflecting the difficulty of reverse tension construction is constructed, including the following levels: The basic indicator layer consists of trajectory parameters obtained through direct collection or calculation, including: Inclination angle; azimuth angle; inclination offset; eccentricity; dogleg; trajectory curvature; rate of change of inclination; rate of change of azimuth; rate of change of offset; The feature index layer, further extracted from the basic indexes, is used to reflect the characteristics of the pilot hole trajectory, including: Trajectory smoothness index; trajectory complexity index; continuous length of abnormal sections index; local anomaly intensity index; continuous high-risk trajectory characteristic index; comprehensive trajectory characteristic index; The construction difficulty characterization layer directly reflects the ease or difficulty and risk level of subsequent reverse tensioning construction, including: Difficulty index of reverse pull; Risk level index; Sensitivity index of difficulty points; Impact index of section anomalies; The difficulty index of reverse pulling can be defined as follows: ; in: The difficulty index of the reverse pull; i = 1, 2, ..., k; k is the total number of local segments. ~ These are the weighting coefficients.

[0011] In the preferred scheme, in S5: based on trajectory characteristics and evaluation index system, an identification and prediction model is established between pilot hole trajectory characteristics and the difficulty level of reverse well reverse pulling construction. The input to the prediction model is the characteristic parameters of the pilot hole trajectory or the evaluation index system constructed from them, and the output is the difficulty level, risk level, and distribution of key and difficult sections of the subsequent reverse tensioning construction; including: Identify the input parameters of the prediction model, including at least three of the following: Average dogleg deviation of the entire hole; maximum dogleg deviation; length of a section with continuous high dogleg deviation; cumulative offset; sudden increase in local offset; curvature variation amplitude; local bending strength; trajectory complexity index; local anomaly strength index; Identify the output of the prediction model, including: Overall difficulty level of reverse tension construction; risk level of each section; location of key and difficult sections; length of key and difficult sections; warnings of key risk sections; The implementation methods for identifying prediction models include one or more of the following: Rule-based decision-making models, weighted evaluation models, mapping relationship models, and statistical or intelligent recognition models.

[0012] In the preferred scheme, in S6: the difficulty level of the reverse tension construction is divided into: Level I: Low difficulty section; Level II: Low to medium difficulty section; Level III: Medium to high difficulty section; Level IV: High difficulty section; Level V: Extremely high difficulty section; The following grading rules: RDI < T1 is Grade I; T1 ≤ RDI < T2 is Grade II; T2 ≤ RDI < T3 is Grade III; T3 ≤ RDI < T4 is Grade IV; RDI ≥ T4 is Grade V; where T1, T2, T3, and T4 are preset grading thresholds; The pilot hole is segmented and analyzed along the well depth direction, and based on the trajectory characteristics of each segment and the model output results, the risk levels of each section are classified into: Low-risk area; Medium-risk area; High-risk area; Extremely high-risk area; Sections with a higher risk level correspond to one or more of the following situations: Sections with continuous high dogleg severity; Sections with continuously increasing offset; Locally bent sections; Sections with concentrated curvature mutations; Sections with superposition of multiple abnormal features.

[0013] In the preferred solution, based on the risk level classification, key difficult sections are further identified, including key control sections that have a significant impact on the safety, continuity, and efficiency of subsequent back-pulling construction; The key difficult sections are identified according to the following principles: The abnormal index of a single trajectory exceeds the preset threshold; Multiple trajectory abnormal features appear concentrated in the same section; The risk level is continuously higher than the preset level; The continuous length of section anomalies is greater than the preset length, and the impact on the drill string force and the stability of construction parameters is greater than the preset value; Through the above identification, construction sensitive sections that need to be key controlled, key monitored, and key warned are screened out from the entire hole range; According to the above analysis and model identification results, one or more of the following results are output: The overall difficulty level of back-pulling construction; The risk levels of each well depth section; The position and length of key difficult sections; Distribution map of high-risk sections; Hints for key difficult section types; List of key attention sections; The output forms include text results, lists, sectional maps, curve graphs, or graphical warning interfaces.

[0014] In the preferred solution, in S7: Construction suggestions based on the identification and prediction results: After identifying and predicting the difficulty of back-pulling construction, risk levels, and key difficult sections, further form subsequent construction suggestions to serve the optimization of construction parameters, risk warning, and construction organization decision-making; 1. For high-risk sections and key difficult sections, give early warning prompts before construction, and it is recommended to focus on monitoring parameter changes such as torque, tension, and penetration efficiency during construction; 2. According to the identification results of high-risk sections and key difficult sections, give suggestions on construction preparation and organization arrangement, including: Specialized technical briefings were conducted for key sections; key personnel were assigned to guard high-risk sections; corresponding emergency response measures were prepared; and a more cautious construction pace was adopted for high-risk sections.

[0015] 3. For extremely high-risk sections, stricter control measures will be implemented to reduce the risks of uneven wear, stuck drills, broken rods, etc.

[0016] 4. Historical engineering samples can be used to train and correct the prediction model, including the characteristic parameters of the pilot hole trajectory and the actual torque, tension, construction efficiency, stuck drill events, downtime events, and tool wear during the corresponding reverse pull stage, providing suggestions for subsequent construction. By introducing historical samples for model training and correction, the model recognition accuracy is further improved, and more targeted suggestions are provided for subsequent construction. This invention not only enables the identification and prediction of construction status, but also allows the identification and prediction results to be directly transformed into engineering application guidance.

[0017] A reverse well reverse pull construction difficulty prediction system is used for a reverse well reverse pull construction difficulty prediction method, including a control unit, a detection unit and an output unit, wherein the control unit is signal connected to the detection unit and the output unit; The control unit includes a data preprocessing module, a trajectory feature extraction module, an evaluation modeling module, and a construction difficulty identification and risk level classification module. The detection unit includes a data acquisition module; The output unit includes a result output module.

[0018] In the preferred embodiment, the data acquisition module is used to collect the hole depth, inclination angle, azimuth angle, trajectory coordinates, deviation, dogleg degree, and construction process parameters during the pilot hole drilling process; The data preprocessing module is used to perform noise reduction, alignment, completion, and standardization on the raw data; The trajectory feature extraction module is used to extract features such as skewness, dogleg degree, rate of change of curvature, offset, and continuous abnormal segments. The evaluation modeling module is used to establish an evaluation model for the difficulty of reverse tension construction; The construction difficulty identification and risk level classification module is used to identify high-risk sections and difficulty identification types. The results output module is used to output the difficulty level of the reverse pull and warning information.

[0019] An electronic device includes a processor and a memory, the memory storing a computer program that runs on the processor for a method for predicting the difficulty of reverse well reverse pull construction and / or for a system for predicting the difficulty of reverse well reverse pull construction.

[0020] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used for a method and / or a system for predicting the difficulty of reverse well reverse pull construction.

[0021] This invention provides a method and system for predicting the difficulty of reverse well reverse pulling construction. By adopting the above scheme, the following beneficial effects are achieved: It can effectively utilize the pilot hole trajectory feature information to make advance judgments on reverse pull construction, thus eliminating the need to rely on in-process adjustments and post-event responses to problems, which is conducive to the stable progress of reverse pull construction using the reverse well method.

[0022] By comprehensively analyzing the trajectory characteristic parameters such as well inclination angle, azimuth angle, well deviation, and dogleg degree obtained during the pilot hole drilling process, the trajectory information of the pilot hole can be more fully explored and utilized.

[0023] A comprehensive and accurate analysis of the spatial morphological characteristics and abnormal changes of the pilot hole trajectory is conducted, transforming the pilot hole trajectory data from simple quality verification data into basic data that can support subsequent construction prediction and analysis.

[0024] By preprocessing, extracting features, and constructing indicators from trajectory data, an identification and prediction model is established between the characteristics of the pilot hole trajectory and the difficulty of reverse well pullback construction. This model can identify and predict the construction difficulty, risk level, and key challenging sections before reverse well pullback construction, thereby improving the pertinence and predictability of construction preparation work.

[0025] The construction control approach is shifted from in-process adjustment and post-event handling to pre-event prediction and feedforward control, thereby improving the initiative of construction management. This helps reduce risks such as uneven wear, stuck drills, and broken rods during reverse tensioning construction, improving the safety and stability of reverse tensioning construction.

[0026] The ability to identify challenging and risky sections in advance allows for proactive preventative measures, reducing construction interruptions and temporary adjustments, and improving the continuity and efficiency of reverse-pull construction. Attached Figure Description

[0027] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a diagram of the evaluation system architecture of the present invention. Detailed Implementation

[0028] In an embodiment of this application, a method for predicting the difficulty of reverse well reverse pull construction includes the following steps: S1. Collect trajectory characteristic parameters during the drilling process of the pilot hole in the inclined shaft; During the drilling of the pilot hole in the inclined shaft, trajectory acquisition equipment such as inclinometers, measurement while drilling systems, and trajectory measurement devices are used to collect the trajectory characteristic parameters of the pilot hole; The trajectory characteristic parameters include three or more of the following parameters corresponding to the well depth: Well inclination angle I; azimuth angle A; well inclination offset OI; azimuth offset OA; total angle change rate, i.e., dogleg degree D; trajectory curvature C; offset O; Simultaneously collect auxiliary parameters related to the drilling status, including drilling pressure, rotational speed, torque, feed rate, flushing medium flow rate, and flushing medium pressure, for subsequent identification and optimization correction of the prediction model; The above parameters are collected at predetermined depth intervals or in real time along the drilling direction of the pilot hole, forming the corresponding raw data sequence.

[0029] 1. Single-parameter sequence representation: For any trajectory parameter, its discrete sequence along the well depth direction can be expressed as: X = {X1,X2,X3,…,X} n}; in: X represents a sequence of characteristic parameters of a certain trajectory; X i This represents the parameter value at the i-th measuring point; n is the total number of measuring points.

[0030] 2. Multi-parameter joint sequence expression: In practical applications, various trajectory parameters have corresponding relationships at the same measurement point, and the multi-parameter joint sequence is constructed as follows: S={S1S2,S3,…,S n}; Wherein: S i =(L i ,I i A i O i D i C i ,…); S i =The multidimensional trajectory feature vector of the i-th measurement point; in: L i This represents the well depth or starting position of the measurement segment corresponding to the i-th measurement point; "..." indicates that other parameters can be added.

[0031] 3. Vectorized representation: Unified representation in matrix form: S= ; Right now: S∈R n×k ; in: n is the number of measurement points; k is the parameter dimension (e.g., k=6); n×k represents an n-row, k-column matrix; R (real number space) represents all values ​​that are real numbers (continuous values).

[0032] S2. Preprocess the collected trajectory data; Because field-collected data may contain missing measurements, false readings, fluctuations, and noise, trajectory data preprocessing is necessary before modeling and analysis. Preprocessing includes, but is not limited to, the following beneficial actions on the dataset: 1. Data alignment: All trajectory parameters are arranged uniformly according to well depth, and the correspondence between different parameters at the same measuring point is established.

[0033] 2. Outlier handling: Remove obviously distorted data caused by measurement errors, equipment fluctuations, or abnormal data transmission to avoid unreasonable impacts of local abnormal data on subsequent analysis results.

[0034] 3. Missing value completion: For missing measurement point data, interpolation can be used to complete the data, so that the trajectory parameters are continuously distributed in the well depth direction.

[0035] 4. Standardization process: Trajectory parameters of different dimensions should be standardized or normalized for subsequent comprehensive evaluation and model input. (Range criterion can be used.) After preprocessing, a standardized trajectory dataset suitable for subsequent feature extraction and modeling is obtained.

[0036] S3. Obtain preprocessed trajectory data and extract trajectory features that characterize the overall spatial morphology and local anomaly changes of the pilot hole; Based on the preprocessed data, trajectory features of the pilot hole are extracted to reflect its overall spatial morphology and local anomaly characteristics. These trajectory features include overall trajectory features, local anomaly features, and comprehensive trajectory features, among which: 1. Overall trajectory feature extraction, used to characterize the overall orientation and smoothness of the pilot hole, including: Average rate of change of inclination angle of the entire well ; Average value of the azimuth change rate of the entire hole ; Distribution of deviation distance in the entire wellbore; Full-hole azimuth offset distribution; Average dogleg degree of all holes ; Maximum dogleg degree of full hole ; Average curvature of the entire hole ; Total Hole Offset Cumulative ; in: Average dogleg degree across all holes: , i=1,2,…,n; Average curvature of the entire hole: , i=1,2,…,n; Average rate of change of inclination angle of the entire well: ; Average rate of change of azimuth angle throughout the borehole: ; Total hole offset cumulative amount: , i=1,2,…,n; in: n is the number of test segments; This represents the offset of the i-th measuring point relative to the design axis. Let i be the dogleg degree of the i-th measurement point; Let be the curvature of the i-th measurement point; This represents the offset of the i-th measuring point relative to the design axis. Let be the well inclination angle of the i-th measuring point; Let be the azimuth angle of the i-th measuring point.

[0037] 2. Local anomaly feature extraction, used to identify local trajectory problems that are detrimental to subsequent reverse tensioning construction, including: Locally high dogleg sections; Sections with sudden increases in local azimuth offset; Local curvature abrupt change section; Local eccentricity abrupt change section; Continuous length of abnormal sections; Sections where the well inclination or azimuth angle changes drastically; A segment can be identified as an abnormal segment if it meets any of the following conditions: ; ; ; ; ; in , , , , The preset threshold; 3. Comprehensive trajectory feature extraction facilitates the construction of recognition and prediction models, and further enables the development of comprehensive trajectory feature indicators, including: (1) The smoothness index of the whole hole trajectory is used to comprehensively characterize the continuity, uniformity and stability of the trajectory change of the pilot hole in the whole hole range, and to reflect whether the whole drilling trajectory is smooth and straight, and whether there are many sharp bends and sudden changes. The larger this index is, the smoother the overall borehole trajectory; the smaller this index is, the greater the fluctuations and the more local abrupt changes in the overall borehole trajectory. The formula can be defined as: ; in, The smoothness index of the entire hole trajectory; These are weighting coefficients, and all are greater than 0; The standard deviation of the full-hole dogleg tolerance; The standard deviation of the curvature of the entire hole; The greater the average dogleg angle, the more bends there are overall, and the less smooth the ride. The greater the maximum dogleg angle, the more pronounced the sharp bends in certain areas, and the less smooth the ride. The larger the standard deviation of doglegity and the standard deviation of curvature, the more uneven the variation and the lower the smoothness. (2) Trajectory complexity index, which is used to comprehensively reflect the geometric complexity of the pilot hole trajectory over the entire hole range, and to reflect whether the trajectory changes frequently, whether the spatial shape is complex, and whether the direction adjustment is drastic; The larger this indicator is, the more complex the trajectory is, and the more complex the stress state of the drill bit is likely to be during the subsequent reverse pull operation. The calculation formula is: ; in: This is a metric for trajectory complexity. These are the weighting coefficients; This represents the number of abnormal segments; If the dogleg degree is high, the curvature is large, the orientation and inclination change frequently, and there are many abnormal segments, the complexity will increase significantly. (3) Local anomaly intensity index, used to characterize the intensity of trajectory anomaly changes in a local segment, focusing on reflecting the concentration of problems such as local high dogleg, high curvature abrupt change, and sudden increase in offset; This indicator is suitable for identifying key and difficult sections as well as local high-risk sections; The calculation formula is: ; in, Let be the local anomaly intensity index for the i-th segment; These are the weighting coefficients; For the i-th local segment, the following can be selected: The average dog-leg degree of this section ; Maximum dog-leg degree in this section ; The curvature change amplitude of this section ; The amplitude of the offset change in this section ; The larger the local anomaly intensity index, the more concentrated and severe the trajectory anomaly in that section is, and the more likely it is to become a construction difficulty during subsequent pull-back; (4) Trajectory offset cumulative index, used to characterize the overall cumulative degree of deviation of the actual trajectory of the pilot hole relative to the design axis, reflecting the cumulative effect of the whole hole deviating from the design axis; The larger this indicator is, the more obvious the overall deviation of the trajectory from the design axis is, and the greater the potential adverse impact on subsequent reverse tensioning construction. Calculation formula: ; in This is a trajectory offset indicator; Let the first The offset of each measuring point relative to the design axis is The length of the measured section is This is the total length of the hole; (5) Continuous high-risk trajectory characteristic index, used to characterize the length and severity of the section in the pilot hole where high-risk trajectory characteristics appear continuously, highlighting the impact of "continuous anomaly" on subsequent reverse pull construction; In engineering, it is not often the single point of abnormality that is most dangerous, but rather a continuous stretch of poor trajectory that is more likely to cause problems such as uneven wear, stuck drill, and broken rod; Calculation formula: ; in: This is a characteristic indicator of a continuous high-risk trajectory; This is the total length of the hole; Assume there are a total of The first consecutive high-risk zone The length of each high-risk section is Its average anomaly intensity is ; It is formed by the combination of the average dogleg degree, average rate of change of curvature, and average rate of change of deviation within the high-risk section; defined as: ; The above steps can reflect the overall spatial shape of the pilot hole and identify abnormal sections that may have an adverse impact on the reverse tensioning construction.

[0038] S4. Construct an evaluation index system that reflects the difficulty of reverse tension construction; To further construct an evaluation index system that reflects the difficulty of reverse tension construction, including the following levels: 1. Basic indicator layer, consisting of trajectory parameters obtained through direct collection or calculation, including: Inclination angle; azimuth angle; inclination offset; eccentricity; dogleg; trajectory curvature; rate of change of inclination; rate of change of azimuth; rate of change of offset; 2. Feature index layer, further extracted from basic indicators, is used to reflect the characteristics of the pilot hole trajectory, including: Trajectory smoothness index; trajectory complexity index; continuous length of abnormal sections index; local anomaly intensity index; continuous high-risk trajectory characteristic index; comprehensive trajectory characteristic index; 3. Construction difficulty characterization layer, used to directly reflect the difficulty and risk level of subsequent reverse tensioning construction, including: Difficulty index of reverse pull; Risk level index; Sensitivity index of difficulty points; Impact index of section anomalies; The difficulty index of the reverse pull is defined as follows: ; in: The difficulty index of the reverse pull; The smoothness index of the entire hole trajectory; This is a metric for trajectory complexity. The average local anomaly intensity of the entire hole is defined as: , i=1,2,…,k; k is the total number of local segments; This is a cumulative indicator of trajectory offset. Characteristic indicators of continuous high-risk trajectories; ~ These are the weighting coefficients.

[0039] The aforementioned evaluation index system transforms the pilot hole trajectory characteristic parameters into evaluation quantities that characterize the difficulty and risk level of reverse-pull construction.

[0040] Evaluation system architecture diagram as follows Figure 1 As shown, the evaluation index system includes, from bottom to top: Basic indicator layer → Feature indicator layer → Construction difficulty characterization layer → Construction decision layer.

[0041] S5. Establish an identification and prediction model between the pilot hole trajectory characteristics and the difficulty of the inverse pulling construction method. Based on the trajectory characteristics and the evaluation index system, establish an identification and prediction model between the pilot hole trajectory characteristics and the difficulty of the inverse pulling construction method: The input of the identification and prediction model is the pilot hole trajectory characteristic parameters or the evaluation index system constructed from them, and the output is the difficulty level, risk level of the subsequent inverse pulling construction, and the distribution of key difficult sections; including: 1. Model input, including at least 3 of the following: Average dogleg severity of the whole hole; Maximum dogleg severity; Length of the locally continuous high dogleg severity section; Cumulative offset; Local offset sudden increase value; Curvature change amplitude; Local bending strength; Trajectory complexity index; Local anomaly strength index; 2. Model output, including: Overall inverse pulling construction difficulty level; Risk level of each section; Location of the key difficult section; Length of the key difficult section; Hint of the key risk section; 3. Model implementation methods, using one or more of the following methods: (1) Rule judgment model, judge the difficulty and risk level of the inverse pulling construction according to whether the trajectory characteristic parameters exceed the preset threshold and the combination of multiple indicators; (2) Weighted evaluation model, assign different weights to multiple trajectory characteristic parameters, construct a comprehensive difficulty index, and divide the construction difficulty level according to the size of the comprehensive difficulty index; (3) Mapping relationship model, form the mapping relationship between the trajectory characteristic and the construction difficulty by corresponding analysis of the pilot hole trajectory characteristic and the inverse pulling construction performance in the historical project; (4) Statistical or intelligent identification model, under the condition of sufficient sample data, regression analysis, classification identification or intelligent learning methods can also be used to establish a more adaptable identification and prediction model.

[0042] S6. According to the identification and prediction model, identify and predict the difficulty, risk level and key difficult sections of the inverse pulling construction; According to the identification and prediction model, conduct difficulty identification and division of the inverse pulling construction difficulty level for the subsequent inverse pulling construction: 1. Based on the whole hole trajectory characteristics and the inverse pulling difficulty index RDI, divide the inverse pulling construction difficulty level of the entire inclined shaft inverse well inverse pulling construction into: Level I: Low difficulty section; Level II: Medium-low difficulty section; Level III: Medium-high difficulty section; Level IV: High difficulty section; Level V: Extremely high difficulty section; The following grading rules: RDI < T1, it is Level I; T1 ≤ RDI < T2, it is Level II; When T2 ≤ RDI < T3, it is Grade III; When T3 ≤ RDI < T4, it is Grade IV; When RDI ≥ T4, it is Grade V.

[0043] Among them, T1, T2, T3, and T4 are preset classification thresholds; 2. Conduct segmented analysis on the pilot hole along the well depth direction, and based on the trajectory characteristics of each segment and the model output results, divide the risk levels of each section, including: Low-risk area; Medium-risk area; High-risk area; Extremely high-risk area; The sections with higher risk levels correspond to one or more of the following situations: Continuous high dogleg severity sections; Sections with continuously increasing offsets; Locally bent sections; Sections with concentrated curvature mutations; Sections with superposition of multiple abnormal characteristics.

[0044] Through the above division, the risk spatial distribution in the subsequent back-pulling construction is obtained.

[0045] On the basis of the risk level division, further identify the key difficult sections, including the key control sections that have a significant impact on the safety, continuity, and efficiency of the subsequent back-pulling construction; The key difficult sections are identified according to the following principles: The single trajectory abnormal index exceeds the preset threshold; Multiple trajectory abnormal characteristics appear concentrated in the same section; The risk level is continuously higher than the preset level; The abnormal continuous length of the section is greater than the preset length, and the impact on the drill string force and the stability of construction parameters is greater than the preset value; Through the above identification, the construction sensitive sections that need to be key controlled, key monitored, and key warned are screened out from the whole hole range; According to the above analysis and model identification results, output one or more of the following results: The overall back-pulling construction difficulty level; The risk levels of each well depth section; The location and length of the key difficult sections; The distribution map of high-risk sections; The prompt of key difficult section types; The list of key attention sections; The output forms include text results, lists, sectional maps, curve graphs, or graphical warning interfaces.

[0046] S7. Output the identification and prediction results, and form targeted construction suggestions.

[0047] Construction suggestions based on the identification and prediction results: After identifying and predicting the back-pulling construction difficulty, risk level, and key difficult sections, further form subsequent construction suggestions to serve the optimization of construction parameters, risk warning, and construction organization decision-making; 1. For high-risk sections and key difficult sections, give early warning prompts before construction, and it is recommended to focus on monitoring the changes of parameters such as torque, tension, vibration, and penetration efficiency during construction; 2. Based on the identification results of high-risk sections and key and difficult sections, suggestions are made for construction preparation and organization, including: Specialized technical briefings were conducted for key sections; key personnel were assigned to guard high-risk sections; corresponding emergency response measures were prepared; and a more cautious construction pace was adopted for high-risk sections.

[0048] 3. For extremely high-risk sections, stricter control measures will be implemented to reduce the risks of uneven wear, stuck drills, broken rods, etc.

[0049] 4. Use historical engineering samples to train and correct the prediction model, including the characteristic parameters of the pilot hole trajectory and the actual torque, tension, construction efficiency, stuck drill events, downtime events, and tool wear during the corresponding reverse pull stage, to provide suggestions for subsequent construction. By introducing historical engineering samples for model training and correction, the model recognition accuracy is further improved, and more targeted suggestions are provided for subsequent construction. This invention not only enables the identification and prediction of construction status, but also allows the identification and prediction results to be directly transformed into engineering application guidance.

[0050] Through the above technical solution, the present invention can make full use of the trajectory feature parameters obtained during the drilling of the pilot hole in the inclined well to analyze the overall spatial morphology and local abnormal changes of the pilot hole, and establish an identification and prediction model between it and the difficulty of the reverse well method reverse pull construction. This enables the identification and prediction of the difficulty, risk level and key difficult sections of the reverse pull construction, providing a basis for subsequent construction parameter optimization, risk warning and construction organization decision-making. In this way, the reverse pull construction risk can be identified in advance and fed-forward controlled, improving the safety, continuity and construction efficiency of the reverse pull construction, and reducing construction risks such as wear, stuck drill and broken rod.

[0051] In addition, based on the above method, the present invention also provides a reverse well reverse pull construction difficulty prediction system, including a control unit, a detection unit and an output unit, wherein the control unit is signal connected to the detection unit and the output unit. The control unit includes a data preprocessing module, a trajectory feature extraction module, an evaluation modeling module, and a construction difficulty identification and risk level classification module. The detection unit includes a data acquisition module; The output unit includes a result output module.

[0052] The data acquisition module is used to collect data on hole depth, inclination angle, azimuth angle, trajectory coordinates, deviation, dogleg degree, and construction process parameters during the pilot hole drilling process. The data preprocessing module is used to perform noise reduction, alignment, completion, and standardization on the raw data; The trajectory feature extraction module is used to extract features such as skewness, dogleg degree, rate of change of curvature, offset, and continuous abnormal segments. The evaluation modeling module is used to establish an evaluation model for the difficulty of reverse tension construction; The construction difficulty identification and risk level classification module is used to identify high-risk sections and difficulty identification types. The results output module is used to output the difficulty level of the reverse pull and warning information.

[0053] The present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the computer program implements the steps of the above method when it runs on the processor.

[0054] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method.

[0055] This application is not only applicable to inclined shaft construction, but can also be applied to other similar hole-forming constructions.

[0056] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A method for predicting the difficulty of reverse well construction, characterized by: Includes the following steps: S1. Collect trajectory characteristic parameters during the pilot hole drilling process; S2. Preprocess the collected trajectory data; S3. Obtain preprocessed trajectory data and extract trajectory features that characterize the overall spatial morphology and local anomaly changes of the pilot hole; S4. Construct an evaluation index system that reflects the difficulty of reverse tension construction; S5. Establish an identification and prediction model for the characteristics of the pilot hole trajectory and the difficulty of reverse well reverse pulling construction; S6. Based on the identification and prediction model, identify and predict the difficulty, risk level and key challenging sections of the reverse tension construction. S7. Output the identification and prediction results and generate targeted construction suggestions.

2. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 1, characterized in that: in In S1: Trajectory characteristic parameters are collected during the drilling of the pilot hole in the inclined shaft, including: Well inclination angle I; azimuth angle A; well inclination offset OI; azimuth offset OA; total angle change rate, i.e., dogleg degree D; trajectory curvature C; offset O; The trajectory feature parameters are collected at predetermined depth intervals or in real time along the drilling direction of the pilot hole, and the corresponding raw data sequence is formed.

3. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 1, characterized in that: In S3: Based on the preprocessed trajectory data, the trajectory features of the pilot hole are extracted. These features include overall trajectory features, local anomaly features, and comprehensive trajectory features, among which: Overall trajectory characteristics include: average rate of change of inclination angle throughout the well. Average value of the azimuth change rate of the entire borehole ; Overall borehole deviation distribution; Overall borehole azimuth deviation distribution; Overall borehole average dogleg deviation Maximum dogleg width of the full hole Average curvature of the entire hole Total Hole Offset Cumulative Amount ; Local anomaly features include: local high dogleg sections; local sections with sudden increases in azimuth offset; local sections with sudden changes in curvature; local sections with sudden changes in eccentricity; continuous length of anomaly sections; and sections with drastic changes in well inclination or azimuth. Comprehensive trajectory characteristics include: full-hole trajectory smoothness index Trajectory complexity index Local anomaly intensity index Trajectory Deviation Cumulative Index Continuous high-risk trajectory characteristic indicators .

4. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 3, characterized in that: in In S4: Construct an evaluation index system reflecting the difficulty of reverse tension construction, including: The basic indicator layer includes: trajectory feature parameters; The feature index layer includes: trajectory smoothness index; trajectory complexity index; continuous length of abnormal segments index; local anomaly intensity index; continuous high-risk trajectory feature index; and comprehensive trajectory feature index. The construction difficulty characterization layer is used to directly reflect the difficulty and risk level of subsequent reverse tension construction, including: reverse tension difficulty index; risk level index; difficulty sensitivity index; and section anomaly impact index.

5. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 4, characterized in that: The difficulty index of counter-pull is defined as: ； in: The difficulty index of counter-pull The average local anomaly intensity of the entire hole is defined as: i = 1, 2, ..., k; k is the total number of local segments. ~ These are the weighting coefficients.

6. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 4, Its characteristics are: In S5: Based on the evaluation index system, an identification and prediction model is established for the pilot hole trajectory characteristics and the difficulty level of the reverse well method reverse pull-out construction. Identify the input parameters of the prediction model, including at least three of the following: Average dogleg deviation of the entire hole; maximum dogleg deviation; length of a section with continuous high dogleg deviation; cumulative offset; sudden increase in local offset; curvature variation amplitude; local bending strength; trajectory complexity index; local anomaly strength index; Identify the output of the prediction model, including: Overall difficulty level of reverse tension construction; risk level of each section; location of key and difficult sections; length of key and difficult sections; warnings of key risk sections; The identification prediction model uses one or more of the following: Rule-based decision-making models, weighted evaluation models, mapping relationship models, and statistical or intelligent recognition models.

7. The method for predicting the difficulty of reverse well construction according to claim 1, Its characteristic is that, in S6, the difficulty level of reverse tension construction is divided into: Level I: Low difficulty section; Level II: Low to medium difficulty section; Level III: Medium to high difficulty section; Level IV: High difficulty section; Level V: Extremely high difficulty section; The grading rules are as follows: RDI < T1 is Grade I; T1 ≤ RDI < T2 is Grade II; T2 ≤ RDI < T3 is Grade III; T3 ≤ RDI < T4 is Grade IV; RDI ≥ T4 is Grade V; where T1, T2, T3, and T4 are preset grading thresholds; The pilot hole is segmented and analyzed along the well depth direction, and based on the trajectory characteristics of each segment and the model output results, the risk levels of each section are classified into: low-risk area; medium-risk area; high-risk area; extremely high-risk area.

8. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 7, characterized in that: Based on the risk level classification, identify the key difficult sections; the key difficult sections include the key control sections that have a significant impact on the safety, continuity, and efficiency of the subsequent back-pulling construction; Screen out the construction sensitive sections that need to be key-controlled, key-monitored, and key-alerted from the entire hole range, and identify the key difficult sections according to the following principles: The abnormal index of a single trajectory exceeds the preset threshold; multiple trajectory abnormal characteristics appear concentrated in the same section; the risk level is continuously higher than the preset level; the abnormal continuous length of the section is greater than the preset length, and the impact on the drill string force and the stability of construction parameters is greater than the preset value; Output one or more of the following identification and prediction results: The overall back-pulling construction difficulty level; the risk levels of each well depth section; the position and length of the key difficult sections; the high-risk section distribution map; the key difficult type prompt; the list of key attention sections.

9. The method for predicting the difficulty of reverse well reverse pulling construction according to claim 8, characterized in that: in In S7: After obtaining the identification and prediction results, further form subsequent construction suggestions, including: For high-risk sections and key difficult sections, give early warning prompts before construction, and it is recommended to focus on monitoring the changes of parameters such as torque, tension, and penetration rate during construction; According to the identification results of high-risk sections and key difficult sections, put forward corresponding suggestions for construction preparation and organization arrangement; Use historical engineering samples to train and correct the prediction model.

10. A system for predicting the difficulty of reverse well construction, characterized in that: For a reverse well method back-pulling construction difficulty prediction method according to any one of claims 1-9, including a control unit, a detection unit, and an output unit, the control unit is signal-connected to both the detection unit and the output unit; The control unit includes a data preprocessing module, a trajectory feature extraction module, an evaluation modeling module, and a construction difficulty identification and risk level classification module The detection unit includes a data acquisition module; The output unit includes a result output module; The data acquisition module is used to collect the hole depth, hole inclination angle, azimuth angle, trajectory coordinates, deviation, dogleg severity, and construction process parameters during the pilot hole drilling process; The data preprocessing module is used to perform denoising, alignment, complementation, and standardization processing on the original data; The trajectory feature extraction module is used to extract the deviation, dogleg severity, curvature change rate, offset, and continuous abnormal section features; The evaluation modeling module is used to establish a back-pulling construction difficulty evaluation model; The construction difficulty identification and risk level classification module is used to identify the high-risk sections and difficulty identification types of back-pulling; The result output module is used to output the back-pulling difficulty level and early warning information.

Images