The integrity of natural gas pipelines will decrease with an increase in operating time, thus causing pipeline leaks and accidents. However, it is challenging to improve the precision and automation of existing sensors to raise leak prediction and classification precision. Therefore, based on deep learning, a 1D convolutional neural network (CNN) incorporating the channel attention mechanism is proposed for recognizing and classifying the type of natural gas pipeline leakage. Firstly, the data reconstruction of the leaked acoustic signals, which have been classified by energy modes, is performed by feature augmentation and Bessel filtering. Subsequently, a lightweight CNN is proposed, and an attention mechanism is introduced to optimize the model performance. The results show that the training performance of the network with the attention mechanism is superior to that of the original network and the network with batch normalization. The attention mechanism network is then used to train the leakage signals with different features of engineering parameters. Finally, the test accuracy achieves 97.81%, validating the effectiveness of the proposed method for identifying and classifying natural gas leaks. It presents new ideas for the implementation of deep learning in the natural gas and chemical industries.

In natural rock mass, the existence of joints or fissures makes the strength of surrounding rock significantly reduced. In this paper, the influence of deformation and failure law of jointed rock mass on the stability of surrounding rock is studied by prefabricating five kinds of single fracture rock samples with different angles. Firstly, the failure and deformation of the unanchored and anchored specimens under different loading conditions (uniaxial compression and cyclic loading) were studied in laboratory tests, respectively. Secondly, the microscopic fracture characteristics and deformation damage modes of the specimens were simulated by using PFC numerical simulation software. The results show that the peak strength and elastic modulus of the sample increase with the increase in the inclination angle. After the sample is anchored, the peak strength of the sample is effectively improved, and the number of cracks is significantly reduced. In practical engineering applications, bolts can change the crack propagation mode and limit the initiation of cracks in the surrounding rock. By studying the crack propagation law and failure characteristics of fractured anchored rock samples, we can deepen the understanding of rock fracture behavior and better reveal the failure and deformation mechanism of surrounding rock instability in tunnel construction.

玛湖油田为典型的致密砾岩油藏,常用"密切割、长水平段、水平井+体积压裂"的开发方式进行投产,整体具备三线性流模型特征,但砾岩基质独特的复模态结构,导致其又与常规的三线性流模型有所差异.因此,从砾岩油藏的复模态结构出发,分析其基质渗流特征,同时以压裂水平井的三线流模型为基础,建立考虑基质复模态结构下的渗流数学方程,并利用换元变换、Pedrosa变换、摄动变换等方法联合对数学模型求解后进行了流态划分、敏感性参数分析及现场应用.研究表明:受改造区基岩中基质与未改造区供液双重影响,本文模型双线性流特征更加明显;同时储容比、窜流系数、启动压力梯度及应力敏感对曲线影响阶段不同,其中储容比、应力敏感越大,压力曲线局部越向上偏移,窜流系数、启动压力梯度越大,压力曲线局部越向下偏移.矿场应用表明:该模型与实际数据拟合较好,符合储层实际情况,可为砾岩储层压裂水平井的动态监测提供一定的指导意义.

In order to study the main control factors of volcanic reservoirs with complex oil–water relationships, the Carboniferous in the Chepaizi Uplift of the Junggar Basin was taken as an example and the lithofacies characteristics, main control factors, and hydrocarbon accumulation model of volcanic reservoirs were investigated by combining the petroleum geology with field testing (data of core analysis, well logging, formation testing, and production testing). The results show that the Carboniferous in the Chepaizi Uplift experienced three stages of volcanic activities and developed seven volcanic lithofacies bodies, distributed in a bead-string connected planar form along the Hongche fault. There is no unified oil–water interface across the whole study area and there are multiple oil–water systems within one fault block. The Carboniferous volcanic reservoir experienced two stages of hydrocarbon accumulation from two different source rocks. The distribution of faults penetrating hydrocarbon kitchens and source rocks controls the macro-scale distribution of reservoirs. The physical properties of reservoirs affect the pattern of oil and water differentiation in volcanic rock bodies, while the lithofacies body-controlled hydrocarbon accumulation mode highlighting “one rock body for one reservoir” determines the distribution of reservoirs. The matching between the paleo-structure and hydrocarbon accumulation stage controls the accumulation and adjustment of hydrocarbon distribution. The Permian source rocks in the Shawan Sag serve as the lateral hydrocarbon supply and hydrocarbons accumulate in the Carboniferous structural-lithologic traps, which are summarized as the two stages of hydrocarbon accumulation of newly generated hydrocarbons into older reservoir rocks. This study of the hydrocarbon accumulation pattern in volcanic rocks aims at guiding the development of Carboniferous reservoirs with complex oil and water relationships in this area.

玛湖地区油气资源丰富,已发现三级地质储量超10×108t,是新疆油田公司油气上产最现实的主战场.针对开发前期存在储层非均质性强、储量丰度低、效益建产难等开发难题,通过创新运用非常规储层改造理念,推行全生命周期一体化管理办法,开展勘探开发一体化、地质工程一体化、地面地下一体化、技术经济一体化、试验建产一体化联合攻关研究,采用非常规开发油藏技术经济指标优化技术,达到提高单井产量和采收率、降低工程投资和运行成本的目的.通过一体化实践成功运用,玛湖地区已建产能超700×104 t,年产量达到200多万吨,取得较好的开发效果.

针对致密油藏压裂水平井流态复杂及全生命周期产量预测误差大的问题,在对前人经验递减模型适用性研究的基础上,选取以不稳定流和过渡流为主导的产量指数递减(SEPD)模型及以边界控制流为主导的指数模型进行组合,通过结合节点处的产量相等和递减率相等,构建了以边界控制流时间为节点的全生命周期新型分段产量预测模型,同时建立了广义回归神经网络算法下的边界控制流时间预测方法及分段函数的最小二乘法拟合参数求取方法.结果表明,不论是否到达边界控制流,新模型普遍比单一SEPD模型和指数模型的拟合度高,其预测结果更接近于生产后期的指数评估结果,误差小于5％.

There exists a large horizontal principal stress difference (11–38 MPa) in the tight conglomerates in the Mahu oil area, China. It is difficult to form a complex fracture network via hydraulic fracturing under these conditions. To improve reservoir stimulation, the fracture formation mechanism of the complex conglomerate fracture networks was explored. Based on the geomechanics theory of fracture formation, the mechanism of the “stress wall” formed by fracturing in horizontal wells was analyzed in this paper. The inhibitory effect of the stress wall on the formation of tensile and shear fractures was studied. The reason for the decrease in the stress difference coefficient caused by fracturing fluid was analyzed through numerical simulation, which suggested that the complexity of a fracturing network is mainly controlled by the interference of externally applied stress and the reduction in the coefficient of internal stress difference. In this paper, innovative technologies were developed by proactively introducing stress interference in the application of the Ma131 small-well-spacing pilot area. The core technologies include optimization of the 3-D staggered small-well-spacing pattern, and synergetic optimization of multiple elements and zipper fracturing. The positive effects of proactive stress interference on improving fracturing volume, reserve utilization rate and recovery were discussed. Based on the concept of proactive stress interference, the “serial fracturing mode” of horizontal wells was proposed to reduce drilling and fracturing interference and improve the development effect.

The conglomerate matrix and fracture propagation are special in tight conglomerate reservoir with small well spacing. In this article, the fractal propagation characteristics of the fracture network in conglomerate reservoirs are described by experiment and a micro-mathematical model. According to the core slice, the conglomerate reservoir matrix presents the multi-modal pore structure, described as the “pseudo-dual-media” model. Given the above, the unsteady seepage mathematical model, comprehensively considering the fractal fracture network, stress sensitivity of main fractures, and threshold pressure gradient of the reservoir matrix, was developed and analytically solved. The Blasingame type curves for production decline analysis were plotted, and the sensitive parameters were analyzed. The field application was performed for validation. The research results show that the fractal dimension decides the complexity of the fracture network distribution. As it increases, the unsteady flow occurs earlier, and the boundary flow is delayed. The anomalous diffusion exponent represents the smoothness of crude oil migration and a higher value leads to higher resistance to oil migration and larger pressure drawdown for the same production rate. The growth of the threshold pressure gradient within a certain range can result in a localized downward shift of the type curves. The field application in a conglomerate oil reservoir showed that the presented model presents a fitting accuracy 10% higher than that of the conventional SRV model and has high reliability and precision for the production performance evaluation of the small-well-spacing development of tight conglomerate reservoirs.

Studying the pore structure of hydrate sediments is helpful to explore the growth and distribution of hydrates. However, it is difficult for a single fractal model to describe the full-scale pore space of sediments. In this paper hydrate sediments are collected from the South China Sea for routine petrophysical measurements. The fractal dimensions ofhigh-pressure mercury intrusion (HPMI) and low-temperature nitrogen adsorption (LTNA) are calculated and compared. Fractal dimension Db (average 2.0203) of small pores based on HPMI cannot represent the small pore heterogeneity. The large pore fractal dimension Da (average 2.6825) calculated from HPMI and the small pore fractal dimension Dc (average 2.521) calculated from LTNA can characterize the fractal characteristics of the entire pore. Compared with shales and tight sandstones, sediments have the lowest pore heterogeneity. The increase in permeability is only related to the increase in the complexity of the large pores. Due to the dissolution of carbonate, the complexity of macropores is reduced. The increase in the content of illite increases the pore complexity of the entire pore size. It is of great significance to evaluate the pore structure of sediments based on the combined fractal analysis of HPMI and LTNA.

储层岩石力学和地应力特征是井眼稳定性分析与压裂设计等的基础,为油气钻采工程提供依据.致密砾岩储层岩石非均质性较强、岩石物理响应较复杂,尚无成熟、有效的测井评价方法实现储层岩石力学和地应力的精准评价.通过岩石力学测试,揭示了致密砾岩储层岩石的动态弹性模量、静态弹性模量及岩石力学强度之间的关系.在此基础上将实验所得关系与测井数据相结合,建立砾岩储层岩石动静态弹性参数转换关系,将测试数据与压裂施工数据结合建立了水平地应力测井计算方法.最终形成了一套适用于致密砾岩储层的岩石力学和地应力测井评价方法,并通过实际应用验证了方法的有效性.

Research on in situ stress has important theoretical and practical significance for the exploration and development of oil and gas reservoirs. The orientation and magnitude of in situ stress in the Gaoshangpu Oilfield northern area (GO-NA) were analyzed using borehole breakout data and acoustic emission measurements. Mechanical experiments, logging interpretation, and seismic data enabled spatial characterization of rock mechanics parameters. A 3D geological model and 3D heterogeneous rock mechanics field of the GO-NA were constructed. Petrel and ANSYS modeling provided detailed prediction of the 3D stress field in the GO-NA. The results indicate that the maximum horizontal stress orientation in the GO-NA is generally ENE–WSW-trending, with significant changes in in situ stress orientation within and between fault blocks. Along surfaces and profiles, stress magnitudes are discrete and in situ stress is of the Ia-type. Observed inter-strata differences were characterized by five different types of in situ stress profile. Faults are the most important factor in the large distributional differences in the stress field of reservoirs observed within the complex fault blocks, significantly affecting magnitudes and orientations in the stress field. The next most important influence on the stress field is the reservoir’s rock mechanics parameters, which affect in situ stress magnitudes. A strong linear correlation exists between reservoir depth and in situ stress magnitude. This technique provides a theoretical basis for more efficient exploration and development of low-permeability reservoirs. It also serves as a reference for the detailed prediction of inter-well in situ stress in regions with similarly complex fault blocks.

Field experience shows that horizontal well multi-stage fracturing technology can transform the volumetric pressure crack network in the formation and has been widely used in unconventional oil and gas resource exploitation. According to the requirements of horizontal well fracturing operations, this paper simulates the law of horizontal well fracturing in low-porosity and low-permeability sand-shale formation by large-mold fracturing tests and analyzes crack initiation and extension. The two completion methods of open-hole completion and casing perforation completion are considered, and the maximum and minimum horizontal principal stress difference is 10 and 4 MPa. The horizontal wellbore azimuth angle changes from 0° to 90°. The test results show that the principal stress difference has a great influence on fracture initiation. The smaller the stress difference, the higher the fracture initiation pressure and the more complicated the fracture initiation and extension. The fracture propagates along the axial direction of the wellbore, and then at both ends of the wellbore the fracture turns to the direction perpendicular to the direction of the minimum horizontal principal stress. The horizontal wellbore can form a transverse fracture by drilling along the direction of the minimum horizontal principal stress. This study is of great significance for understanding the fracture initiation and propagation law of horizontal well fracturing and can provide guidance for multi-stage fracturing operations of horizontal wells in unconventional reservoirs.

After fracturing operations, a large amount of fracturing fluid is retained in shale fracture network, resulting in low flowback efficiency. This has been attributed to the imbibition of fracturing fluid into matrix pores. However, it is unclear how the imbibition mechanism is involved, what are its governing laws and controling parameters in fracture networks? Based on the three-dimensional water imbibition theory of matrix blocks, a fracture network model is established, and a number of dimensionless controling parameters are proposed and analyzed for flowback efficiency. The results show that the imbibition characteristics of fracturing fluid in fracture network are mainly determined by two dimensionless numbers; namely, dimensionless imbibition time, fracture width, and imbibition capacity. The dimensionless imbibition time characterizes the contact time between the fracturing fluid and shale formation, which negatively correlates to the flowback efficiency. The dimensionless fracture width is the ratio of the fracture width to the rock length, which is inversely proportional to the flowback efficiency. Smaller value of the dimensionless fracture width corresponds to larger contact area of fracturing fluid and shale, leading to a lower flowback efficiency. The dimensionless imbibition capacity depicts the capacity of shale reservoirs to imbibe fracturing fluid, which has a negative linear correlation with flowback efficiency. In addition, dimensionless time and fracture width are related to the fracturing operations, and are enhanced by increasing the shut-in periods and proppant concentration. Therefore, the flowback efficiency can be controlled by changing fracturing operations. The predictive method of the flowback efficiency established here is of great significance for reservoir damage analysis and flowback regime optimization. Cited as : Yang, L., Wang, S., Cai, J., Gan, Y., Salama, A. Main controlling factors of fracturing fluid imbibition in shale fracture network. Capillarity, 2018, 1(1): 1-10, doi: 10.26804/capi.2018.01.01