Actual Drilling Data Research Articles

Three-pressure prediction method of jointing well-seismic data in JT1 well area of Sichuan Basin in China

With the high yield of many wells represented by Well JT1 in the Maokou Formation, has catalyzed a surge in exploration activities along the platform margin facies of the Maokou Formation in central Sichuan and further showed the significant exploration potential of the Maokou Formation in the northern slope. However, the fracture cave body of the Maokou Formation exhibits a high degree of development, strong longitudinal and horizontal heterogeneity, large formation pressure differences, and drilling events such as gas kicks and lost circulation occur frequently, which seriously affects the efficient implementation of drilling. Understanding the spatial distribution of the three-pressure in the formation can help better deal with and solve the above problems. Therefore, in order to help the safe, high-quality and rapid drilling of the Maokou Formation in the study area, and enhance the efficiency of oil and gas development, this paper explores the research on the prediction method of the three-pressure of jointing well-seismic data based on the geomechanical experimental data and the actual drilling data. In the process of prediction of pore pressure, this study found that the pore pressure and formation velocity in the study area have an exponential relationship. In order to enhance the applicability of the Filippone’s method in the study area and improve the prediction accuracy of pore pressure, the linear relationship between pore pressure and formation velocity in the Filippone’s method is modified to an exponential relationship, and a pore pressure prediction model suitable for the work area was established. Based on the Mohr–Coulomb criterion and Huang's model, the prediction models of collapse pressure and fracture pressure applicable to the study area were established, respectively. Then, the elastic parameters were obtained through pre-stack inversion, and the three-pressure bodies were calculated based on the elastic parameter bodies. The results indicate that: (1) The three-pressure prediction method of the jointing well-seismic data in this paper can predict the formation's longitudinal and transverse pressure anomaly zones in advance. (2) The Maokou Formation in the study area is characterized by abnormally high pressure, to balance the pressure of the high-ground formation, high-density drilling fluid is necessary. (3) The prediction results of three-pressure in this paper are highly consistent with the actual drilling engineering events, which verifies the reliability of the three-pressure prediction results presented in this study. The results of the study can provide a basis for decision-making in drilling geological design, such as the determination of drilling fluid density, the evaluation of borehole stability and other engineering problems that require support from three-pressure data.

Dynamic Numerical Simulation and Transfer Learning-Based Rapid Rock Identification during Measurement While Drilling (MWD)

In constructing rapid rock identification models for measurement while drilling (MWD) via neural network methods, collecting actual drilling data to train the model is extremely time-consuming and labor-intensive. This requires extensive drilling experiments in various rock types, resulting in limited neural network training data for rock identification that covers a limited range of rock types. To suitably address this issue, a dynamic numerical simulation model for rock drilling is established that generates extensive drilling data. The input parameters for the simulations include torque, drill bit rotation speed, and drilling speed. A neural network model is then developed for rock classification using large datasets from dynamic numerical simulations, specifically those of granite, limestone, and sandstone. Building upon this model, transfer learning is appropriately applied to store the knowledge obtained in the rock identification based on the neural network model. Further training through transfer learning is conducted with smaller datasets obtained during actual drilling, making the model suitable for practical rock identification and prediction in the drilling processes. The neural network rock classification model, incorporating dynamic numerical simulation and transfer learning, achieves a prediction accuracy of 99.36% for granite, 99.53% for sandstone, and 99.82% for limestone. This reveals an enhancement in prediction accuracy of up to 22.94% compared to the models without transfer learning.

Prediction method for formation pore pressure based on transfer learning

Pore pressure prediction of drilled wells is the recognition of formation information using logging and other actual drilling data after drilling is completed, which is of great significance in improving the mastery of regional formation information and reducing the potential risks of the engineering program of wells to be drilled. Traditional formation pressure prediction methods rely on manual experience and are difficult to cope with complex formations. Existing machine learning-based methods have high accuracy in the prediction of formation pressure for wells with similar data composition in the same block or logging, but have poor accuracy in the prediction of wells with large differences in data from different blocks or logging. In this paper, a transfer learning algorithm is introduced to modify the existing LSTM-based formation pore pressure prediction model and establish a transfer learning-based formation pore pressure prediction method. In addition, we evaluated the prediction performance of machine learning algorithms and transfer learning algorithms under the same/different blocks and the same/different logging data conditions. The results show that, compared with the existing machine learning algorithms, for the same block and the same logging data conditions, transfer learning can improve the prediction accuracy by 0.15%; for the same block and different logging data conditions, transfer learning can improve the prediction accuracy by 0.33%; for the different block and the same logging data conditions, transfer learning can improve the prediction accuracy by 0.19%; for the different block and different logging data conditions, transfer learning can improve the prediction accuracy by 0.11%. Overall, the addition of the transfer learning method can improve the accuracy and generalization ability of the formation pore pressure prediction model based on machine learning to a larger extent.

A drilling combination positioning method based on geomagnetic/inertial navigation combined with PSD improved unscented Kalman filtering algorithm*

During the drilling process, the accuracy of measuring drilling tool attitude is severely affected by measurement drift errors in the inertial device itself and the significant interference noise generated by the impact vibration of the drill bit. To address these issues and achieve precise measurement of drilling trajectory, this paper proposes a combined positioning method based on geomagnetism/inertial navigation combined with PSD improved unscented Kalman filtering. An experimental platform was constructed, and a sensor system was installed in the vicinity of the drill bit for short section measurements. Experimental verification was conducted using actual drilling data. The results demonstrate that, compared to general unscented Kalman filtering, the average position error using PSD improved unscented Kalman filtering was reduced from 0.122 m to 0.088 m, and the average velocity error was reduced from 0.019 m s−1 to 0.005 m s−1. By comparing and matching with the drilling trajectory drawn in the logging process, the overall positioning accuracy of the combined positioning method is as high as 90.8%, yielding positioning results that are highly authentic and effective, thereby meeting the on-site usage requirements.

Hybrid Approach Using Physical Insights and Data Science for Stuck-Pipe Prediction

Summary Early detection of stuck-pipe incidents is crucial because of the enormous costs of recovering from such incidents. Previous studies have leaned significantly toward a physics-based or data science approach. However, both approaches have challenges, such as the uncertainty of the physics-based model and the lack of data in the data science approach. In this study, we propose a hybrid approach using physical insights and data science and discuss the possibility of stuck-pipe prediction. The proposed method comprises two steps. In the first step, a data-driven model with physical insights is trained using the historical data of the in-situ well to estimate some of the drilling variables. In the second step, the risk of stuck-pipe occurrence (hereafter referred to as sticking risk) is calculated based on the historical and current measured data and the estimation of the trained model. This approach is expected to overcome the limitations of the previous methods as it allows the construction of a detection model tuned to the in-situ well. In the case studies, models for estimating the topdrive torque and standpipe pressure were constructed. The performance of the models is discussed using actual drilling data from drilling fields, including 21 stuck-pipe incidents during drilling operations. The proposed method was first examined using short-term output. The output confirmed that the sticking risk increased shortly (up to 20 hours) before the stuck-pipe incident occurred in 15 cases. This increase in sticking risk was consistent with physical considerations. Subsequently, this study examined the long-term output over several months; this was rarely done in previous studies. Even within this long-term output, some cases had good performance with only a few false alarms, while others had problems with many false alarms. For cases of low performance, several model improvements, such as adding mud information or making the learning and threshold-setting methods more robust to outliers, were found to have the potential to improve performance. The novelty of our research lies in creating a broad framework for the stuck-pipe prediction by using both physical insights and data science methods. The proposed hybrid approach demonstrated the potential to reduce false alarms and improve interpretability compared with previous methods. The framework is highly extensible, and further performance improvements can be expected in the future.

Periods and Processes of Oil and Gas Accumulation in the HZ-A Structure Double Paleogene Field, Pearl River Mouth Basin

The source of oil and gas and the stages of oil and gas accumulation in the “double-Paleo” field of the HZ-A structure in the Pearl River Mouth Basin are analyzed, and the spatiotemporal coupling relationship of the key conditions of oil and gas accumulation are discussed to reconstruct the process of oil and gas accumulation. Based on previous research results, which are based the characteristics of biomarker compounds, the oil and gas in the HZ-A structure double Paleogene field came from the Paleogene Wenchang Formation hydrocarbon source rocks in the HZ26 sub-sag. By means of the casting thin section identification and inclusion homogenization temperature measurement, this paper reveals the three major hydrocarbon accumulation periods and corresponding fluid charging types in the “double-Paleo” field of the HZ-A structure in the Pearl River Mouth Basin. The results show that 13.8–10 Ma is the charging period of low mature crude oil, 10–5.3 Ma is the charging period of mature crude oil, and from 5.3 Ma is the natural gas charging period. Based on actual geological, drilling, logging, and seismic data, the key conditions for hydrocarbon accumulation in the HZ-A structure “double-Paleo” field are sorted out; that is, the source conditions are characterized by high-quality lacustrine source rocks generating early oil and late gas and a near-source continuous hydrocarbon supply. The reservoir conditions are characterized by weathering and superposition of a fracture zone that transforms into a reservoir, and a large-scale sandstone rock mass that transforms into a reservoir. The caprock conditions are characterized by the stacking of several thin mudstones that form a seal and the combination of multiple lithologies that block hydrocarbon migration. The trap conditions are characterized by multistage uplift structure traps and fracture-lithology combination control traps. The transport conditions are characterized by multi-stage cross-bed transport of source-connected faults and lateral differential transport of shallow sand in deep fractures. Finally, oil and gas accumulation models of the HZ-A structure double Paleogene field were established, and the accumulation process was reconstructed. The overall process involved three stages, with the first stage being the localized oil-displacing-water mode, the second being the large-scale oil-displacing-water mode, and the third being the late progressive gas-displacing-oil mode.

Comparative Study of the Seismic Response Characteristics of Three Special Soils

Because of its special material composition, formation environment and special structure, special soil usually shows different seismic response characteristics from general soil, and sometimes causes serious seismic disasters. Therefore, the study of seismic response characteristics of special soil has important practical significance for engineering construction and disaster prevention in special land areas, and a site seismic response analysis is an important way to study the impact of site conditions on ground motions. Based on the existing research work, the uniform and single ideal soil layer profiles of three special soils are established, and the seismic response analysis program of the soilquake soil layer is used to study the effects of the dynamic shear module ratio, damping ratio and shear-wave velocity of soft soil, loess and laterite on the characteristic period and platform value of the design response spectrum under different input ground motion intensities. At the same time, three typical actual drilling data values are selected to establish the calculation model of the site soil layer. On the basis of the seismic response analysis and calculation of the soil layer, three special soil dynamic characteristics are compared from the aspects of peak acceleration and characteristic parameters. Based on the research results of this paper, the recommended values of the dynamic shear module ratio and damping ratio of three special soils are given. This work has important reference value for seismic design in soft soil, loess and laterite areas, and also enriches the research results of soil dynamic parameters.

A Novel Rate of Penetration Model Based on Support Vector Regression and Modified Bat Algorithm

In the geological drilling process, predicting the rate of penetration (ROP) is significantly important for improving drilling efficiency and reducing non-drilling time. However, due to the drilling data pollution and the complex nonlinearity in the geological drilling process, a reliable and highly accurate ROP prediction model is not easy to construct, and the model accuracy is affected by the value of model hyperparameters. In order to overcome the difficulties in modeling, a novel ROP model is developed to handle abnormal data and achieve nonlinear fitting. First, a local outlier factor is introduced to automatically detect the abnormal data, and then replace it with the nearest normal data. Then, the support vector regression (SVR) method is applied to construct nonlinear prediction model for ROP, and a modified bat algorithm (MBA) is developed to solve the non-convex problem in determining optimal value of hyperparameters for SVR-based ROP model. The MBA has six modifications to improve the global search ability, which achieves better performance in global search ability compared with other nine algorithms based on the experiments of IEEE 2005 benchmark functions. The developed ROP prediction model that combines SVR and MBA methods is tested based on actual drilling data and a semi-physical system, and the simulation and application results show that the developed modeling method has higher ROP prediction accuracy compared with the other modeling methods.

Application of long short-term memory network for wellbore trajectory prediction

For improving the accuracy of wellbore trajectory prediction, especially in the build section, a new prediction model based on long short-term memory (LSTM) network was proposed. At the same time, the model was built by Python language and TensorFlow library. The well inclination and azimuth angle were predicted by the LSTM model. Moreover, the prediction accuracy of the LSTM was compared with that of the minimum curvature method. The results show that the average prediction error of the LSTM is much 50% lower than the minimum curvature method, and the proposed model in this article is more consistent with the actual drilling data. This method does not rely on any assumption of the path shapes and geometries. In addition, the model proposed in this article is easy to use and convenient for the engineering field application without the derivation of mathematical formulas.

A novel optimization method for geological drilling vertical well

Geological drilling is an important means for exploration of the earth. Due to the nonlinearity and coupling, geological drilling is accompanied by low efficiency and safety, and difficult to accurately predict the drilling states. Moreover, the vertical well trajectory needs to follow the plumb line. A novel optimization method is proposed to solve these problems. First, working conditions are identified by fuzzy C-means clustering method and corresponding adjustment ranges of operating variables are refined for optimization. Meanwhile, support vector regression and hybrid bat algorithm are employed to construct reliable models for mud pit volume (MPV) and rate of penetration (ROP). Then, how to improve safety and efficiency is converted to a two-objective problem, that is to suppress MPV fluctuations and improve ROP with vertical well constraints. Nondominated sorting genetic algorithm II is invoked to solve this problem, and the appropriate solution is selected by two references points. Moreover, a short-time scale and long-time scale optimization strategy is introduced to cope with different adjustment interval of operating variables. Finally, the simulation results based on actual drilling data, application results in a semi-physical experiment system and comparison results from a vertical well verify the effectiveness and practicability for the developed method.

A Weak SNR Signal Extraction Method for Near-Bit Attitude Parameters Based on DWT

In practice, the near-bit signals are usually accompanied by noises generated by drilling tool vibration and high-speed rotary. This study introduces discrete wavelet transform to a well drilling area and presents a new weal signal extraction algorithm, which can eliminate the vibration and rotary noises effectively and obtain the useful gravity attitude signals based on the discrete wavelet transform through a hard threshold method. The effectiveness and usefulness of our method are verified via Matlab simulation and actual drilling data.

Gas-bearing reservoir distribution in a volcanic environment: Case of the Yingcheng Formation, China

As a special type of lithologic body, gas-bearing strata, composed of rhyolite, an igneous rock with a volcanic composition, are characterized by deep burial, complex lithology, poor stratification, and a complicated gas-bearing distribution. These characteristics lead to difficulties in the prediction of gas-bearing reservoir parameters and thus restrict the exploration and exploitation of such reservoirs. The unique and complex volcanic model contains different types of facies and subfacies because the volcanic eruption usually experiences different explosion-effusion periods. Our study focuses on the vertical and lateral gas-bearing reservoir distribution in a volcanic environment in the Cretaceous Yingcheng Formation of the Songnan Gas Field, using seismic facies-controlled nonlinear stochastic inversion to integrate the volcanic facies information. First, we built the volcanic facies model for the Songliao Basin to introduce the comprehensive characteristics of the volcanic facies during a volcanic eruption and analyzed the well-tie logging facies and seismic facies to guide the facies analysis of the target layer. Then, based on this facies analysis, we depicted the gas-bearing reservoir in detail through the use of seismic facies-controlled nonlinear random inversion. Finally, we predicted the distribution of the gas-bearing reservoir by the inversion results. The predicted results were found to agree well with the actual drilling data. The inversion method effectively predicts the velocity distribution of the gas-bearing reservoir, which provides a good foundation for the subsequent description of gas-bearing reservoirs in the Yingcheng Formation.

Developing a geomechanics-modeling based method for lost circulation risk assessment: A case study in Bohai Bay, China

As a challenging engineering problem, the lost circulation in fractured formations seriously hinders the efficient development of oil and gas wells. Clarifying the weak layer is the key to improve lost circulation prevention and remediation. Nevertheless, due to the complexity, randomness, and interference of engineering factors, a mature lost circulation prediction method of fractured formation has not yet been formed. This work, based on geomechanical modeling, combining with logging and seismic data, constructs the three-dimensional fracture intensity spatial distribution, the three-dimensional minimum in-situ stress spatial distribution, and the three-dimensional brittleness index spatial distribution of the target oilfield, respectively. At the same time, the analytic hierarchy process is used to establish a three-dimensional lost circulation risk index of the target oil field, and a classification evaluation standard for lost circulation risk is proposed. This method is applied in the Bohai A oilfield, the three-dimensional lost circulation risk index of the entire oilfield are established. Three hundred seventy-seven lost circulation sample points of drilled wells are selected, and a classification evaluation standard for lost circulation risk is established for the Bohai A oilfield. The research results show that the classification evaluation standard for lost circulation risk has good guiding significance for subsequent drilling in this area. It can scientifically guide the design of well trajectory and well structure of subsequent wells, and improve lost circulation prevention. • Combining logging data, seismic data, and actual drilling data of Bohai A oilfield, this work makes a clear relationship between rock mechanics environment, rock brittleness, fracture development intensity, and lost circulation. • This work uses geomechanical modeling methods to calculate the spatial distribution of 3D fracture intensity, 3D minimum in-situ stress, and 3D brittleness index of Bohai A oilfield. Further, a three-dimensional lost circulation risk assessment index is constructed based on the analytic hierarchy process. Moreover, according to the lost circulation data of drilled wells, a classification evaluation standard for lost circulation risk in Bohai A oilfield is formulated. • This work comprehensively considers the main factors affecting loss, fully reflects the lost circulation characteristics, and quantitatively predicts the fracture lost circulation zone of the formation.

Analysis on borehole stability of gas drilling with complete stress\u2013strain and thermal stress theory

In conventional mud drilling, sufficient drilling fluid density is required to maintain wellbore stability. However, in gas drilling, the support ability of gas in wellbore is very weak, and the wellbore stability is usually good. This shows that the borehole instability mechanism of gas drilling is different from that of mud drilling. In gas drilling, the borehole wall rock still has a certain bearing capacity after reaching the peak strength. In this paper, the stability of gas drilling wellbore is studied by using the complete stress–strain model. The deformation of rock around borehole in gas drilling can be divided into elastic zone, plastic zone, and broken zone. The radii of plastic zone and broken zone were used to evaluate the stability of borehole wall. This method has a good general applicability to the case of no thermal stress in borehole wall rock. However, due to throttling and cooling as gas flows out of the bit nozzle during gas drilling, thermal stress occurs when the cryogenic flow in the annulus meets the hot formation. Therefore, thermal stress is introduced into the complete stress–strain calculation model. The radii of plastic zone and broken zone are calculated according to the actual drilling data, and compared with the measured borehole diameter. The results show that the model is reasonable and effective. Analysis shows that in gas drilling process, the borehole temperature is much lower than the original formation temperature near the bottom of the borehole, and the borehole temperature is higher than the formation temperature in the upper part of the borehole. Thermal stress has a significant effect on borehole stability. The thermal stress in the bottom borehole makes the rock shrink and offset the expansion of the original in situ stress on the surrounding rock, thus enhancing the stability of the borehole wall. The surrounding rock in the top borehole expands under the action of thermal stress, enhancing the expansion of the surrounding rock toward the borehole and aggravating the instability of the borehole wall. The model can be used to better explain the wellbore instability mechanism of gas drilling, and it is more intuitive to use plastic zone radius and broken zone to characterize wellbore stability of gas drilling.

Early sign detection for the stuck pipe scenarios using unsupervised deep learning

In this paper we present a novel approach for detecting early signs for the stuck events in drilling using Deep Learning. Specifically, we adapt neural network based unsupervised learning tool called Autoencoder for anticipating the ‘stuck’ events during the drilling process. We build Autoencoders on Recurrent Neural Networks (RNNs) to model the normal drilling activity, thereby detecting the stuck incidents as anomalous activity. We conduct experiments on the actual drilling data collected from 30 field wells operated by multiple drilling sources with diverse well profiles and demonstrate that our approach obtains promising results for the stuck sign detection. Furthermore towards explaining the trained model’s prediction, we present reconstruction analysis on the individual drilling parameters.

Prestack Inversion Identification of Dolomite Reservoirs in the Fourth Member of the Sinian Dengying Formation in Moxi Area, Sichuan Basin, SW China

The dolomite reservoir of the fourth member of Dengying Formation in Moxi area of Sichuan Basin is thin, is fast in lateral variation, and has P-impedance difference from the surrounding rock; it is difficult to identify and predict the dolomite reservoir and fluid properties by conventional poststack seismic inversion. Through the correlation analysis of core test data and logging P-S-wave velocity, this work proposed a formula to calculate the shear wave velocity in different porosity ranges and solved the issue that some wells in the study area have no S-wave logging data. AVO forward analysis reveals that whether the gas reservoir of dolomite reservoir is located at the top of the fourth member of Dengying Formation is the main factor affecting the variation of AVO type. Through cross-plotting analysis of elastic parameters, it is found that P-S-wave velocity ratio and fluid factor are sensitive parameters to gas-bearing property of dolomite reservoir in the study area. By comparing the inversion results of prestack parameters such as density, P-wave impedance, S-wave impedance, P-S-wave velocity ratio, and fluid factor, it is found that the gas-bearing prediction of dolomite reservoir by using P-S-wave velocity ratio and fluid factor obtained from simultaneous prestack inversion had the highest coincidence rate with actual drilling data. At last, according to the distribution characteristics of fluid factor and P-S-wave velocity ratio, the favorable gas-bearing areas of dolomite reservoir in the fourth member of Dengying Formation in the study area are finely predicted, and the next favorable exploration areas were pointed out.

Optimization of drilling performance using various metaheuristics

AbstractThe most crucial function in drilling wells is the rate of penetration, which is modeled by many researchers, and the best one is Young-Bourgyen model, which is used in this study. Eight factors affecting rate of penetration have been studied and approved in developing a mathematical equation that shows the combined effects of these variables on rate of penetration optimization. This paper presents an efficient way to find the optimum values for parameters of the Young-Bourgyen model using metaheuristic algorithms. An actual drilling data was used from Khangiran field to calculate the difference between the actual penetration rate and the predicted one by different optimization algorithms. Particle swarm optimization, dynamic differential annealing optimization, artificial bee colony, gray wolf optimization, Harris hawk's optimization, flower pollination algorithm, firefly algorithm, whale optimization algorithm, and sine cosine algorithm are used to find best possible solution.

Analysis of physical properties of middle-deep reservoirs in the east of Jizhong Depression

The Jizhong Depression is one of the hot spots for oil and gas exploration in North China. As the Jizhong Depression has the largest area and richest oil and gas resources, the Raoyang Depression has good prospects for exploration. This paper takes the middle-deep reservoirs of the Shahejie Formation in the Raoyang Depression as the research object. Aiming at the problem of the unclear characteristics of the middle-deep reservoirs in the Raoyang Depression, using core description, thin section identification, and cathodoluminescence technology, the physical properties of the middle-deep reservoirs in Shahejie are clarified. The following points of understanding have been obtained: The rock types of Shahejie Formation in Raoyang area are mainly feldspar sandstone and detrital feldspar, followed by feldspar detrital sandstone and feldspar quartz sandstone. The main mineral content of different structural belts is different. The quartz content in Suning and Maxi areas is higher than that in Liuchu and Dawangzhuang areas. The feldspar content in Liuchu and Dawangzhuang areas is higher than that in Suning, Maxi and Liuxi areas. The Maxi area has the highest lithic content. There are 7 types of pores, including intergranular pores, intragranular pores, granular dissolved pores, mold pores, intergranular dissolved pores, intragranular dissolved pores, and intergranular pores. The porosity in Raoyang area is between 0.85% and 35.26 % with the average value of 12.05%. The permeability is 0.01mD-1312mD, and the average value is 12.75mD. It belongs to low porosity-low permeability reservoirs. Finally, combined with sedimentary facies characteristics and actual drilling and logging data, the reservoir division of the study area is completed, and 10 favorable areas are proposed. The research results provide a reference for the next step in the study of the physical characteristics of the middle and deep reservoirs in the eastern part of the Jizhong Depression.

An online hybrid prediction model for mud pit volume in the complex geological drilling process

The mud pit volume (MPV) model is of great importance in evaluating the bottom hole pressure (BHP). In this paper, an online hybrid model is developed to predict MPV considering the drilling characteristics of data pollution, multi-variable, strong nonlinearity, and time series characteristics. First, the mutual information and fast Fourier transform method are introduced to filter data noises and determine the model inputs. Then, back propagation neural network (BPNN) method and support vector regression (SVR) method are used to establish the submodels, and the submodels are combined based on three evaluation criteria. After that, the combination model is fine-tuned according to the time series trends of MPV based on the long short-term memory neural network (LSTMNN). Finally, a modified sliding window method is developed to update the hybrid model constructed by SVR, BPNN and LSTMNN. The simulation results based on actual drilling data show that the online hybrid model has higher accuracy than other prediction models, and the online hybrid model can follow the time series characteristics of MPV, which validates the effectiveness of the developed model.

Optimization design method for the bypass trajectories of infill adjustment wells in the fracturing areas of shale gas fields

In the design of a shale-gas cluster horizontal well, it is necessary to consider the bypass of the fracturing influence domains of existing wells and the interference between fracturing influence domains when the wellbore trajectories of infill adjustment wells in the fracturing areas are designed. In order to quickly evaluate the rationality of the design scheme of fracturing wellbore trajectory in an infill adjustment well, this paper adopted the vector algebra method to build a geometric model of the obstacles in the shale gas fracturing area. In this geometric model, the influence domains of hydraulic fractures are taken into account. Then, based on this geometric model, the optimization design model of bypass trajectory in the shale gas fracturing area was established by taking the minimization of total trajectory length and trajectory potential energy as the optimization objective and the anti-collision between trajectories as the constraint. Besides, the geometric check method to judge if there is any interference between fracturing influence domains was provided. Finally, the established optimization design model was verified based on the actual drilling data of Fuling Shale Gas Field in the Sichuan Basin. And the following research results were obtained. First, the obstacle sizes in fracturing areas will be seriously underestimated if the fracturing influence domains are neglected. Second, if the fracturing influence domains are neglected, the designed bypass trajectory can bypass the wellbore trajectories of old wells, but may intersect the fracturing influence domains of existing wells, thus inducing drilling accidents. In conclusion, the proposed optimization design model of bypass trajectory in the shale gas fracturing area can satisfy the constraint of anti-collision and bypass and achieve the optimization objective of minimizing total trajectory length and trajectory potential energy, and the corresponding design calculation avoids complex calculation and check.