Accurate prediction of wellbore trajectory is critical for drilling safety and efficiency. However, strong parameter coupling and rapid trajectory variation in complex geological conditions challenge conventional models. Long Short-Term Memory (LSTM) networks are highly sensitive to hyperparameter settings, and fixed configurations often lead to overfitting and a delayed response to dynamic drilling environments. To address this issue, this paper proposes a SAC-LSTM framework integrating the Soft Actor-Critic (SAC) algorithm with LSTM. Unlike traditional offline optimization strategies, SAC enables continuous hyperparameter adjustment within a continuous action space, maintains exploration via entropy regularization, and dynamically adapts parameters using real-time data feedback. Combined with incremental training, the model continuously learns from new measurements. Field data from three wells demonstrate R2 improvements of up to 0.162 (inclination) and 0.129 (azimuth) over conventional LSTM, confirming the superior accuracy and adaptability of the proposed method in real-time trajectory prediction under evolving drilling conditions.

The core tasks of wellbore trajectory control are to control inclination and azimuth of steerable drilling tool. During the drilling process, RSS (Rotary Steerable System) constantly changes the target inclination and azimuth. The non-intelligent downhole closed-loop control method often leads to larger hysteresis of wellbore trajectory control and an increase in non-productive time. This study proposed a control methodology for RSS downhole closed-loop control, which combined a back-propagation neural network with a fuzzy control system (BP-Fuzzy). This paper also investigated the control method of PID, fuzzy, and BP. In the simulation experiments, both inclination and azimuth assigned new targets, and the performance of four control methods were evaluated with a RSS dynamic model. Furthermore, in the simulation, the fuzzy method initializes control parameters using PID values, while the BP-Fuzzy method adopts the same fuzzy rules as the fuzzy method and the same neural network structure as the BP method. Therefore, the simulation experiments are methodically sequential and control other variables. In multiple simulations, BP-Fuzzy method shows better control effect in response speed, overshoot, steady-state error and disturbance resistance. Finally, a three-dimensional drilling trajectory, encompassing vertical drilling, build-up, and horizontal drilling, was planned and implemented, with random disturbance introduced throughout the process. The BP-Fuzzy method exhibited superior performance in tracking the target attitude and demonstrated enhanced disturbance suppression capabilities during the entire drilling operation. This method can be applied to downhole closed-loop control to enhance the automatic performance of RSS and establish the foundation for future autonomous drilling.

[Objective]The azimuthal gamma logging system while drilling is a critical component for achieving precise geological steering.This system uses a shielded crystal within a gamma sensor,allowing only a specific window to detect gamma rays.Furthermore,the system is integrated with attitude measurement technology,and the circumference of the logging point is divided into multiple sectors relative to the gravitational orientation.Based on the division of the sectors,a motor-driven rotating gamma sensor dynamically measures the gamma intensity.Gamma sensors provide sectoral gamma intensity data,facilitating formation comparison and lithological analysis.They can even analyze the borehole trajectory and the relative positions of strata using azimuth gamma analysis,thus providing a basis for geological guidance.However,traditional ground-testing schemes for azimuthal gamma logging systems rely on comprehensive physical testing methods,which result in high costs,low efficiency,and significant safety concerns.[Methods]To solve these problems,this study designed an experimental scheme for azimuthal gamma logging while drilling using semiphysical simulation.First,the mathematical models for the attitude sensor and the gamma-pulse signal sensor were developed based on principles of attitude measurement and azimuthal gamma detection during drilling.The attitude sensor model could output corresponding signals for different spatial attitude angles,such as accelerometer,fluxgate magnetometer,and gyroscope signals;the gamma pulse sensor could generate negative pulse signals with a width of 5 μs for acquisition by the azimuthal gamma logging system while drilling.Based on the developed models,a real-time simulation platform for an azimuthal gamma logging experiment while drilling was established,consisting of an upper computer,an NI-PXI simulator,and an azimuthal gamma measurement system.Among these,the upper computer,designed using the VeriStand software platform,was used for setting the initial model parameters and displaying operational status;the sensor models were deployed and run in the NI-PXI simulator;and the azimuthal gamma logging system while drilling was connected to the simulator for signal transmission,thereby forming a real-time semiphysical simulation platform for the azimuthal gamma logging experiment while drilling.Finally,an automated testing procedure was designed using the VeriStand software,allowing flexible configuration of test duration and initial parameters based on the testing requirements.Three system-level tests were conducted under different initial parameter configurations.The test results included outputs from the sensor models,average gamma intensity,and gamma intensity measurements from specific sectors.[Results]Experimental results demonstrated that the designed semiphysical simulation test system for azimuthal gamma logging while drilling could output dynamic signals for the attitude measurement sensor and azimuthal gamma sensor.The measurement errors of gamma intensity for each sector were less than 5%,with data omission rates below 10 counts per second,meeting the design specifications for the azimuthal gamma logging system and verifying the effectiveness of the automated testing of the designed system.[Conclusions]The designed semiphysical simulation system is capable of simulating the dynamic characteristics of attitude and gamma ray sensors during drilling operations in a surface environment.By eliminating the need for physical sensors and implementing an automated testing procedure,the proposed scheme can significantly reduce test costs,reduce safety risks,and enhance efficiency in testing the azimuthal gamma logging system.

The Sichuan Basin and its periphery are rich in natural gas resources, and unconventional oil and gas reservoirs in igneous/volcaniclastic rocks show great exploration potential. Influenced by the Emei Taphrogenesis, a set of volcaniclastic rocks developed during the deposition of the first member of the Wujiaping Formation (Wu1 member) in the northeastern Sichuan Basin. However, the reservoir types and natural gas accumulation patterns of these volcaniclastic rocks remain unclear. Focusing on well LY1 and well YB701, combined with logging data from multiple exploration wells, core descriptions, geochemical data, and reservoir information, this study clarified the development types and accumulation patterns of natural gas reservoirs in volcaniclastic rocks of the Wu1 member in the northeastern Sichuan Basin. The results showed that: ① The vertical lithological assemblages in the Wu1 member varied greatly across different regions, exhibiting strong heterogeneity, and were generally characterized by interbeds of bioclastic limestone and volcaniclastic rocks. ② The Wu1 member developed two exploration types of reservoirs: "self-generating and self-storing" type and "near-source" type. The "self-generating and self-storing" reservoir type was characterized by high total organic carbon (TOC), high gas content, and high porosity, whereas the "near-source reservoir" type was characterized by low TOC, low gas content, and high porosity. ③ The spatial distribution of the two reservoir types was controlled by the distance from volcanic vents and depositional environment. The reservoir storage space was dominated by clay mineral shrinkage pores, dissolution pores, and microfractures, while organic pores were underdeveloped. The pore structure was dominated by mesopores, followed by micropores, and the preservation conditions were excellent. ④ Two types of natural gas accumulation patterns were developed in the Wu1 member, and there were differences between the two patterns in terms of reservoir spatial distribution and hydrocarbon supply. The Puguang area was characterized by a "sandwich" structure with multiple sources of hydrocarbon supply, whereas the Yuanba area was characterized by a "two-story" structure with a single source of hydrocarbon supply. This study provides guidance and reference significance for the exploration and evaluation of natural gas from volcaniclastic rocks in the Wu1 member.

This article proposes a novel calibration method to improve the calibration accuracy of microelectromechanical system (MEMS) triaxial accelerometers in directional drilling tools. First, a comprehensive sensor error model is established, including scale factor, bias, nonorthogonality, and misalignment. To identify these error factors, a 12-position data collection procedure is designed to acquire raw accelerometer data. Based on this model, a new calibration framework which aims at simultaneously minimizing both spatial attitude angle errors and accelerometer output errors is developed. An improved PID-based search algorithm (PSA) is then employed to solve this optimization problem. Simulation and experimental results demonstrate the superiority of the proposed method. Compared to existing approaches, the maximum output errors are reduced to 0.41%, whereas those of other methods exceed 0.80%. Additionally, the inclination and toolface angle errors are less than 0.1 degrees and 0.3 degrees, respectively, while those of other methods exceed 0.5 degrees and 0.6 degrees, respectively. These results verify the effectiveness and accuracy improvement of the proposed method. With its enhanced performance, this method provides a viable solution for improving attitude measurement for MEMS triaxial accelerometers in directional drilling tools.

When air is entrained into the engine lubricating oil system to form an oil and gas two-phase flow, it will seriously affect the normal operation of the lubrication system. Therefore, it is very important to realize the accurate and rapid measurement of the void fraction of the oil-gas two-phase flow in the engine lubricating oil system. This paper determines the void fraction in the engine lubrication system at the elbow based on the spectral matching method, Firstly, the absorption data of oil-gas two-phase flow were obtained at two flow rates and five temperature conditions, covering a gas content range of 0. 10% to 1.00% (with an interval of 0.06%). This was accomplished by utilizing near-infrared, visible, and ultraviolet spectrophotometers, followed by rigorous analysis, It was ascertained that the oil-gas two-phase flow demonstrates absorption across all three wavelengths, with the intensity of absorption being correlated to the gas content. Secondly, the data preprocessing method combined with spectral similarity measure is proposed and applied to the gas content spectral analysis of bent pipe, significantly reducing the maximum relative error of gas content prediction based on the original spectrum, Using data enhancement methods such as center and autoscaling combined with Spectral Angle Cosine methods in the near-infrared spectrum, the maximum relative error of the gas content of the new lubricating oil was reduced from 48% to 36% relative to the original spectrum, The method predicts the void fraction of gas-oil two-phase flow in three bands respectively. The experimental conditions include two flow rates and five temperatures, and the influence of the temperature and flow rate of the two-phase flow on the void fraction prediction is analyzed. At the temperature of 30.0 C and the flow rate of 5.1 m min, the gas content information in the ultraviolet band (193.5 similar to 413.8 nm) is more closely related to the spectral feature of the direction or shape difference of the spectral vector and the maximum relative error of the gas content prediction is only 6%. In the near-infrared and visible bands, the maximum relative error decreases with the increase of temperature or velocity when the flow rate or temperature is constant. There is no specific effect of temperature on the prediction of gas content in the ultraviolet band, With the increase of the two-phase flow rate, the maximum relative crror of gas content prediction tends to increase. The results show that for new lubricating oil with good light transmittance, the gas content data is collected by the ultraviolet spectrometer, and the maximum relative error of gas content prediction is minimum by using standardized pretreatment combined with spectral Angle cosine method.

Sand production has caused serious harm to the long-term production of oil and gas wells. In order to accurately monitor the sand production, this paper proposes an ANN model based on the improved VMD algorithm to monitor sand mass flow rate. In the monitoring of sand mass flow rate in natural gas pipelines, this paper begins by employing the wavelet time-frequency analysis method to analyze the sand signal. Subsequently, the improved VMD method is utilized to denoise and extract features from the sand signal. Finally, KPCA is applied to reduce the dimensionality of the sand signal features. The results from the laboratory platform demonstrate that the fuzzy entropy, mean square frequency, and peak factor can effectively predict sand mass flow rate as feature inputs to the ISMA-ANN model. This experimental result provides a reliable and efficient method for realtime monitoring of sand mass flow rate in real gas well production.

This study addresses the reliability challenges of motor drive circuits used in rotary steerable systems used under extreme downhole conditions, high temperature, high pressure, and intense mechanical vibration. Dynamic current-sharing imbalance and thermal failure mechanisms inherent to traditional parallel IGBT configurations were analyzed, with emphasis on the divergence caused by parasitic parameters that triggers short-circuit faults and operational failures. To enhance system robustness, an improved drive-circuit architecture based on the SiC MOSFETs is proposed. Through optimized circuit layout and validation under real drilling conditions, the circuit has been demonstrated to significantly enhancement of operational stability and reliability. This study indicates that in extreme downhole environments, simplified architecture utilizing high-performance single-chip devices offers a more reliable technical pathway compared to traditional multi-device configurations.

Deep shale gas has become an important frontier for future shale gas exploration and development. The Wufeng-Longmaxi formations in southern China have undergone complex tectonic and transformation through multi-stage tectonic movements. Deep shale gas enrichment conditions are complex, which greatly restricts the exploration and development of deep shale gas. In this study, based on systematic analysis of basic geological characteristics and gas reservoir characteristics of deep shales, the main factors controlling deep shale gas enrichment in southern China were investigated, and enrichment modes were established. The results show that high-quality shales were developed in the deep-water continental shelf facies, characterized by moderate thermal maturity, high silica content, and abundant organic matter. These characteristics provide a good basis for the formation and enrichment of shale gas. The deep shale gas reservoir is featured by overpressure, high porosity and high gas content. The development and maintenance of high porosity, favorable roof and floor sealing conditions, and weak tectonic activity during uplift stage are the main factors to control deep shale gas enrichment. Based on a comprehensive analysis, the enrichment modes of deep shale gas under three different tectonic patterns were established, namely overpressure enrichment within the basin, overpressure enrichment in the faulted nose or slope of the margin, and overpressure enrichment in the remnant syncline outside the basin. This study provides a reference for exploration and development of deep shale gas in Sichuan Basin and other areas.

A novel in-field calibration method for micro-electromechanical system (MEMS) triaxial accelerometers was proposed to simplify the data collection procedure and improve the calibration accuracy in this article, considering cross-axis sensitivities without the need for external equipment. MEMS accelerometers were placed at six positions on a simple platform, in contrast to the at least 12 positions required by other in-field calibration methods. To enhance calibration accuracy, the cross-axis sensitivities were introduced into the sensor error model comprising 18 calibration parameters. The proposed calibration method was verified by simulations and real experiments, and the errors of scale factor, zero offset, misalignment factor, and cross-axis sensitivity are within 0.002%, 0.07 mg, 4x10(-4) , and 8x10(-4) , respectively. Compared with the existing in-field calibration methods, the Proposed method's output modulus errors are less than 0.51% whereas the others exceed 0.80%. The test results demonstrate that the attitude angle error obtained by the new method is less than 0.2 degrees, confirming the effectiveness of the proposed method. The new method offers higher calibration accuracy and does not depend on external high-precision equipment, making it suitable for in-field calibration of MEMS triaxial accelerometers.

Aiming at the problem that it is difficult to accurately predict wellbore trajectory under complex geological conditions, the NOA-LSTM-FCNN prediction method for steering drilling wellbore trajectory is proposed by combining NOA, LSTM and FCNN. This method adopts LSTM layer to receive input data and capture long-term dependencies within the data, extracting important information. The FCNN layer performs nonlinear mapping on the output of the LSTM layer and further extracts relevant features to enhance prediction accuracy. NOA is employed for hyperparameter optimization of the LSTM-FCNN model. The experimental results show that the prediction effect of the proposed method is better than that of other methods. Taking the prediction results of the well deviation angle of H21 as an example, compared with traditional machine learning methods (LR, SVM and BP) and deep learning methods (CNN, LSTM and GRU), the evaluation index R² of this method was improved by 0.17887, 0.03129, 0.0259, 0.00054, 0.00032 and 0.00031 respectively, showing significant prediction accuracy advantages and strong adaptability. In addition, it applies to various types of wellbore trajectory data, effectively enhancing wellbore trajectory prediction capabilities under complex geological conditions.

The attitude angles of the drilling tool serve as crucial information for transmitting Measurement While Drilling (MWD) data, enabling the optimization of drilling performance and ensuring tool safety. However, the real-time transmission and processing of attitude data pose a significant challenge, especially with the increasing prevalence of horizontal and directional drilling. To accurately and promptly obtain the attitude data, this paper proposes a lossless compression method based on Huffman coding, called Adaptive Frame Prediction Huffman Coding (AFPHC). This approach leverages the slowly varying characteristics of MWD tool attitude data, employing frame residual prediction to reduce data volume and selecting optimal bit widths for encoding transmission data. By using real-world drilling data, the proposed method is implemented on a Verilog HDL on a Xilinx field-programmable gate array (FPGA) circuit. Simulation and experiment results show that compression ratios provided by the proposed method for the inclination, azimuth, and toolface angles reach up to 4.02 times, 3.98 times, and 1.48 times, respectively, outperforming several existing methods.

ObjectiveThe traditional manual control mode of a rotary steerable system (RSS) faces challenges, including response delay and low control efficiency. To address these challenges, this study proposed a closed-loop automated attitude control method for the RSS and verified its effectiveness through hardware-in-loop (HIL) simulation. MethodsThis study focuses on the drilling process of a point-the-bit RSS. In the case of deviations between measured and target attitudes, the control parameters were calculated based on the deviation signals using a closed-loop controller, driving the RSS actuator to conduct trajectory correction. Afterward, the actual attitude was yielded and then entered the input end of the RSS through the feedback of the measurement unit, thereby forming a closed-loop control structure. To further validate the effectiveness of the proposed control algorithm, this study developed a HIL simulation platform consisting of (1) a digital twin model that integrated a model describing the movement of the drilling tool and a virtual attitude sensor model, (2) a real-time simulator, and (3) a physical main control unit of the RSS. This platform enables real-time interactions among the control algorithm, the movement model of the drilling tool, and the virtual attitude sensor model. In this manner, the high-precision simulation of complex downhole conditions can be achieved in a laboratory environment. Using the HIL simulation platform, this study conducted tests and experiments on manual and automated control modes of drilling, along with anti-disturbance tests on the holding interval under the control of analytical formulas and proportional integral derivative (PID). Results and Conclusions Compared to manual control, the automated control mode exhibited significantly reduced overshoot, steady-state errors, and settling time. For the inclination interval, the automated control mode reduced the overshoot from 23.92% to 1.96% and the average settling time by 72.33% compared to the manual control model. For the azimuth interval, the automated control mode reduced the overshoot from 23.31% to 2.04% and the average settling time by 53.45%. To verify the robustness of the automated attitude control method, anti-disturbance tests were conducted under azimuthal disturbances with a mean of zero and a variance of 0.2. The results reveal that the automated attitude control method, combined with PID control, reduced the settling time by 18.9% reduction in the holding section. Therefore, the proposed closed-loop automated control method can significantly enhance the attitude control performance of the RSS, the stability of drilling trajectories, and response efficiency, providing a feasible technical pathway for achieving the automated RSS control under downhole conditions. The digital twin-based HIL simulation platform offers an efficient environment for verifying the iterative optimization of the downhole closed-loop control algorithm, effectively avoiding the high cost and risks associated with tests in the actual downhole environment. This accelerates the iterative optimization of the downhole closed-loop automation algorithm of the RSS and its transformation to engineering implementation.

In practice, the near-bit drilling tool confronts with strong vibrations and high-speed rotation. Therein the original signal amplitude of the tool attitude measurements is relatively feeble, and the signal-to-noise ratio (SNR) is exceptionally low. To handle this issue, this paper proposes a weak SNR signal extraction method, frequency selecting complementary ensemble empirical mode decomposition, which is based on ensemble empirical mode decomposition combining with complementary noise and frequency selecting. This method firstly adds different positive and negative pairs of auxiliary white noise to the original near-bit weak SNR signal, secondly adopts empirical mode decomposition on each pair of noise-added signals, then performs ensemble averaging on the obtained multiple sets of intrinsic mode function (IMF) to output more stable IMF of each order and set suitable weights according to designed frequency threshold, and finally reconstructs the original useful signal through weighted summing IMFs. Simulation results show that the extraction accuracy of well inclination angle ranges about ± 0.51°, and the extraction accuracy of tool face angle ranges about ± 1.35°, and meanwhile experimental results are provided compared with other advanced methods, which verifies the effectiveness of our method.

The errors can accumulate over time when microelectromechanical system (MEMS) accelerometers are in use, it is necessary to perform autocalibration to remove the systematic error effect. However, little research has been done on simultaneously considering the non-orthogonality and misalignment effect in MEMS triaxial accelerometer calibration. To enhance the calibration accuracy and extend the autocalibration methods, this paper proposed a new autocalibration method for MEMS triaxial accelerometers base on a PID-based search algorithm. Non-orthogonality and misalignment error factors were integrated into the error model and a PID-based search algorithm was developed to identify the optimal error factors, minimizing the difference between the actual accelerometer output and the local gravity. The proposed calibration method was tested through simulations, with results showing that the errors in scale factor, bias, non-orthogonality, and misalignment are within 1.5865×10-4, 2.0060×10-4, 5.0621×10-4, and 6.8104×10-4, respectively. The proposed method was also compared with the existing calibration methods, with the maximum errors of the proposed method within 0.41% and those of the others exceeding 0.80%. Simulation and comparison results verified the effectiveness and improvement of the proposed method. The average time of the autocalibration is 13.22 s, offering practical value for MEMS triaxial accelerometers.

To address accelerated failure and aging of key electronic circuits during drilling, a compact signal acquisition module has been designed for monitoring in limited underground space. Utilizing the dsPIC33EV128GM104 microprocessor and signal conditioning circuits, it processes measured signals and converts them with internal AD circuitry. The MCP2551 serves as the CAN transceiver for communication, allowing the module to transmit data to the host computer for real-time signal detection and status feedback, enhancing intelligent monitoring through high-speed and reliable data transmission.

This paper designs a management system for while-drilling circuit board test data based on a MySQL database. The system is a unified Qt interface management tool that organizes and shares while-drilling circuit board test data. It uses a client/server (C/S) architecture, which enhances system response speed and separates data.The system includes functionalities for data uploading and downloading, as well as for creating, deleting, querying, modifying, and visualizing the data. Through integrated system construction and design, the system can structure and unify the management and analysis of the large volumes of data generated by while-drilling capacitor boards, ultimately facilitating easier maintenance and significantly improving security.

As an urgent equipment for unconventional oil and gas development, Rotary Steerable System (RSS) plays an increasingly important role in oil and gas exploitation and other industrial fields. At the same time, in order to improve the drilling efficiency of tools and reduce the error interference caused by human operation, RSS needs to develop in the direction of downhole closed-loop automatic drilling and intelligence. This work focuses on downhole automatic inclination-holding drilling of RSS. Based on the Rotary Steerable System tool, a drilling tool simulation model is first established. Then, by using the deviation between the current inclination and the target inclination, the toolface angle is controlled, enabling the tool to drill or maintain along the target inclination. The control strategy used is fuzzy PID control method, and the control effect is better than the traditional PID control, which provides a new method for RSS intelligent control.

The relay protection sensitivity is one of the determined factors in the power system, however, it is often overlooked in current distribution network (DN) planning. The relay protection sensitivity can be decreased to below the minimum values, failing to meet the requirements for electrical installations. To address this challenge, a new optimization model integrated with the relay protection sensitivity to maximize the inverter interfaced distributed generator (IIDG) penetration level while minimizing IIDG investment was proposed in this paper. The IIDG effect on the relay protection sensitivity was analysed and the relay protection sensitivity re-evaluation method was developed. The relay protection sensitivity evaluation was integrated into the proposed model and the particle swarm optimization (PSO) algorithm was developed to solve the nonlinear issue. The proposed optimization method was tested on different cases, and results confirmed the effectiveness of the method. Furthermore, the relay sensitivity profiles obtained through the proposed method and the optimization without considering the relay sensitivity limits were compared. The proposed method improves the average and minimum sensitivity factors by 28.77 % and 51.76 %, respectively, when the DTO protection functions as the backup for the protected line in the thirty-three-node system. When DTO acts as the backup for the adjacent line, the average and minimum values increase by 29.91 % and 50.95 %, respectively. Comparative analysis confirms the efficacy of the proposed method. The new method extends the power system panning approaches and can be integrated into the DN planning tools to support the low-carbon initiatives.

Drilling motors are widely used in unconventional oil and gas exploration. Due to the increased non-productive time and drilling costs brought about by accidental damage to drilling motors, predictive maintenance for drilling motors is necessary to optimize asset utilization. However, service companies face significant challenges in achieving predictive maintenance: operational data acquisition, automated statistics analysis, and drilling state recognition. This paper presents a miniature vibration recorder, an automatic statistical analysis method, and a layered recognition algorithm to resolve these challenges and improve tool maintenance efficiency. The designed recorder can be installed in the catch of a conventional mud motor to record drilling dynamics over a drilling motor’s entire operation cycle. Time-series data from the recorder can be used to automatically generate operation statistics, mitigating the costs incurred by manual data analysis. The layered recognition algorithm then enables the automatic identification of drilling operation states, i.e., surface, downhole non-drilling, downhole sliding, and downhole rotation. The solutions were validated by deploying the recorder in drilling field runs and analyzing recorded data using the associated design software, yielding a functional data collection, automatic data statistical analysis, and operation state recognition accuracy of 95%. Through achieving improved data collection and analysis, the recorder and software introduced in this work can notify motor owners of the detailed operation history of their tools and enable informed preventive maintenance.