The reservoir in the Shunbei field is characterized by ultra-deep, ultra-high temperature, and ultra-high pressure. During the drilling process, the circulating temperature at the bottom of the well is higher than the temperature resistance of downhole instruments, which leads to frequent problems of device burnout and no signal. Therefore, it is of great significance to accurately predict the wellbore temperature field of ultra-deep directional wells. In this paper, the influence of the drilling string assembly, the flow channel structure and the flow pattern on the convective heat exchange coefficient is considered. Based on the energy conservation equation, a numerical model of wellbore-formation transient heat transfer is developed. Then, the model was verified by the real data of two ultra-deep wells in Shunbei block, China, and the results showed that the prediction errors of bottom-hole temperature were all within 2%. Finally, the key factors and rules of the wellbore annulus temperature are analyzed. The results show that the bottom-hole temperature decreases with the decrease of inlet temperature, the thermal conductivity of drilling fluid, and the thermal conductivity of drill pipe, and increases with the decrease of flow rate, the density of drilling fluid, viscosity of drilling fluid, and specific heat capacity of drilling fluid. The inlet temperature has the greatest influence on the outlet temperature, and the specific heat of the pipe string has a minor influence on the wellbore annulus temperature. The research results of this paper provide an accurate wellbore temperature field prediction method for ultra-deep directional wells in the Shunbei block, China, which is of great significance for temperature-controlled drilling.

The timing of the data is not taken into account by the majority of risk warnings today. However, identifying temporal fluctuations in data, which is a vital method for detecting risk, is neglected by the majority of intelligent gas kick warning models now in use. To accurately and early detect the gas kick risk, a temporal series gas kick detection method based on sequence-to-sequence depth autoencoder is proposed in this paper. A depth autoencoder model based on bidirectional long short-term memory (BiLSTM-AE) network is established to encode and compress input series, and decode and reconstruct the output series. Firstly, the BiLSTM-AE network is trained on normal drilling data based on unsupervised learning. Then, the model is tested by gas kick data, and the mean square error of reconstruction is calculated. The results show that the BiLSTM-AE model is more robust and generalized, and its accuracy is 95%. Experimental preliminary results show that this approach is capable of extracting bidirectional temporal information from risk sequence data, but long short-term memory (LSTM) and autoencoder models based on multilayer perceptron (MLP-AE) are unable to do so. By taking into account the temporal characteristics of the data, this study offers a strategy to integrate prior knowledge and significantly enhances the accuracy and stability of the model.

Deep and offshore drilling operations face complex geological formations, uncertain formation pressures, and narrow safety density windows, making them susceptible to lost circulation risks. To address these challenges, this paper introduces an innovative, intelligent lost circulation monitoring model that incorporates geological lithology information. This model not only utilizes real-time drilling parameters, but also encodes geological information such as rock type as inputs to the model. By combining these key lithological features, the model can comprehensively assess wellbore stability and reduce the lost circulation risks. In this paper, the Conditional Tabular Generative adversarial network (CTGAN) model is used to enhance the data of small-sample risk data, which can effectively expand the data distribution space and improve the performance of the model. This paper conducts a comparative analysis of intelligent monitoring results using artificial neural networks (ANNs), long short-term memory (LSTM), and temporal convolutional networks (TCNs). The results show that the TCN achieves an identification accuracy of 93.7%. Furthermore, the analysis reveals that the inclusion of lithology information significantly enhances the model’s performance, resulting in a 7.1% increase in accuracy. The false alarm rate of the model can be reduced by 10.2%, considering the fluctuation of the logging curve caused by the on/off condition of the pump. This indicates that the introduction of lithology information and the condition of the pump on−off provide advantages in monitoring and identifying lost circulation risks, enabling a more precise assessment of wellbore stability and a reduction in lost circulation incidents. The method of lost circulation monitoring proposed in this paper provides an important safety guarantee for the oil drilling industry.

This study introduces a three-dimensional supervoxel segmentation method to accurately separate solid and fluid phases in X-ray images of porous materials, with applications in energy research. Compared with intelligent segmentation algorithms requiring model training, the proposed method operates as a ready-to-use solution with significantly enhanced efficiency. When benchmarked against conventional approaches such as watershed transformation, our technique demonstrates superior segmentation accuracy. Tested on porous rock and gas diffusion layers under varying wettability, it accurately quantifies fluid saturation, interfacial area, curvature, and contact angles—key parameters for enhanced oil recovery, CO2 storage, and hydrogen fuel cells. The proposed three-dimensional segmentation method is noise-resistant and annotation-free, improving both the accuracy and efficiency of segmenting diverse micro-structural material datasets and providing reliable measurements of their geometric characteristics. Document Type: Original article Cited as: Ye, S., Song, X., Ma, Z., Gao, Y., Zhu, L., Zhou, M., Xiao, L., Wen, G., Bijeljic, B., Blunt, M. J. A noise-resistant and annotation-free supervoxel-based algorithm for rapid segmentation of multiphase X-ray images. Advances in Geo-Energy Research, 2025, 16(1): 50-59. https://doi.org/10.46690/ager.2025.04.06

During the drilling of ultra-deep wells, gas kick often occurs, influenced by the complex void pressure profile. The accurate description of particle settling behavior in the gas–liquid mixture is of great significance to effectively deal with gas kicks and ensure drilling safety. In this study, the gas–liquid two-phase annulus flow with different gas volume fractions is created through the transparent annular pipe, constant pressure air pump, and gas flowmeter. High-speed photography is used to record and analyze the sedimentation of particles in gas–liquid mixtures. This study is based on 288 tests. The main parameters in this experiment include the particle Reynolds number, the gas fraction, and liquid viscosity. The effects of wall and gas fraction on the drag coefficient were analyzed. The correlation of particle terminal settling velocity was established. The results obtained show a correlation with average absolute errors (AAE) of 10.7%. This study reveals the settling characteristics of particles in the annular gas–liquid mixed flow, provides an accurate terminal settling velocity prediction explicit formula, and provides guidance for the calculation of bottom hole pressure under the condition of gas kick.