Hasil untuk "Petroleum refining. Petroleum products"

Sangwon Jang, Taekyung Ki, Jaehyeong Jo et al.

Modern video generators still struggle with complex physical dynamics, often falling short of physical realism. Existing approaches address this using external verifiers or additional training on augmented data, which is computationally expensive and still limited in capturing fine-grained motion. In this work, we present self-refining video sampling, a simple method that uses a pre-trained video generator trained on large-scale datasets as its own self-refiner. By interpreting the generator as a denoising autoencoder, we enable iterative inner-loop refinement at inference time without any external verifier or additional training. We further introduce an uncertainty-aware refinement strategy that selectively refines regions based on self-consistency, which prevents artifacts caused by over-refinement. Experiments on state-of-the-art video generators demonstrate significant improvements in motion coherence and physics alignment, achieving over 70\% human preference compared to the default sampler and guidance-based sampler.

JIE Zhijun, MA Xiaojing, CAI Wenzhe et al.

The tight sandstone gas reservoir in a certain area of Shanxi Province is characterized by low porosity and low permeability. The relationship between pore structure, resistivity and gas and water is complex. The reservoir characteristics pose challenges to the traditional gas layer interpretation and evaluation methods mainly based on resistivity logging. Therefore, it is urgent to seek a gas layer evaluation method that is suitable for such complex reservoirs and well conditions. This paper is based on the small-diameter pulsed neutron saturation logging tool - reservoir evaluation tool (RET), and uses the Fast Neutron Cross Section (FNXS) method to evaluate gas layers.. Based on the feature that FNXS is sensitive to gas but not to non-gas, the FNXS value characterization formula is obtained through Monte Carlo numerical simulation under complex pore structure conditions. The correctness of the FNXS characterization formula is verified in the standard calibration well, and the gas saturation of the reservoir is calculated. The accuracy of this method is verified through field gas testing. Meanwhile, through the data analysis of multiple wells in the block, the FNXS value characterization formula for mudstone correction suitable for this block is established, further improving the accuracy of gas saturation calculation and forming the preliminary gas layer evaluation standard for this block. Practice has proved that the FNXS method based on the RET instrument can quickly identify gas layers under complex formation and well conditions. After slime emendation, the FNXS value characterization is more accurate and the calculation of gas saturation is more precise. This research effectively guides the exploration and evaluation of tight gas in the study block. In - depth research on the FNXS method shows that this method can also be popularized and applied in conventional gas - bearing reservoirs.

ZHU XIANGYU, YU WENQUAN, ZHANG JIANWEI et al.

The deep oil and gas exploration area serves as a crucial position for resource development in Subei Basin . However, challenges including generally poor physical properties of deep reservoirs, insufficient understanding of oil and gas enrichment mechanisms, and ineffective reservoir prediction to meet exploration demands have constrained the expansion of deep oil and gas exploration. To understand the enrichment mechanisms of deep oil and gas, develop key exploration technologies, and indicate future research directions, this paper focuses on the deep layers of Gaoyou and Jinhu Sags, which are rich in oil and gas resources. Firstly, by analyzing the exploration development trends and oil and gas resource potential in oil and gas enrichment Sags such as Gaoyou and Jinhu, along with physical characteristics and main controlling factors of deep reservoirs, it was believed that the deep oil and gas reservoirs in Gaoyou and Jinhu Sags were mainly characterized by low to extra-low porosity and permeability. Secondary pore was the main pore type, while primary pore occurred locally. Overall, as burial depth increased, the proportion of primary pores gradually decreased. Subsequently, based on the relationship between pores and pore throats, deep reservoirs were classified into four types of pore-throat structures: large intergranular pores and wide lamellar throats; small intergranular pores and narrow lamellar throats; intragranular dissolution pores and narrow lamellar throats; and micropores and tubular throats. The physical properties of deep reservoirs were generally poor, with locally developed favorable reservoirs. The factors influencing the physical properties of deep reservoirs were complex. Analysis suggests that sedimentary factors, diagenesis, tectonic activity, oil and gas injection, and abnormal formation pressures all significantly affected the physical properties of deep reservoirs, although the controlling factors and their effects varied across different regions. Secondly, investigations were conducted on the occurrence conditions, main controlling factors, and accumulation models of deep oil and gas. The occurrence conditions of oil and gass suggested that oil and gas migration and accumulation were controlled by the pressure systems and physical properties between source rocks and reservoirs, as well as between different reservoirs. Oil and gas accumulation occurred when migration forces overcame migration resistance. Microscopically, pore-throat structure determined the fluid occurrence state and permeability. Larger throat radii, lower pore-throat radius ratios, and smaller tortuosities led to enhanced pore-throat connectivity and higher reservoir permeability. Macroscopically, pressure increase with oil and gas generation provided the driving force for oil and gas migration and accumulation. The magnitude and direction of source-reservoir pressure difference decided the favorable trends for oil and gas migration and accumulation, controlling their favorable areas. In terms of the main controlling factors for oil and gas enrichment, it was believed that oil and gas accumulation and enrichment in deep reservoirs were jointly controlled by source-reservoir configuration, pressure increase with oil and gas generation, fault-sandstone carrier system, and reservoir physical properties. Three accumulation models for deep oil and gas enrichment were established: stepped accumulation driven by combined abnormal overpressure and buoyancy, accumulation via fault-sandstone carrier system driven by abnormal overpressure, and accumulation of early-stage oil and gas injection followed by later-stage compaction. These models elucidated the enrichment mechanisms of deep oil and gass. Based on the above, to address exploration challenges such as unclear reservoir distribution, undefined enrichment zones, and low identification accuracy of effective reservoirs, three breakthrough technologies were developed: (1) A facies-controlled index method for deep reservoir classification was developed based on “facies-controlled index, porosity-permeability characteristics, pore structures, and diagenetic facies”. Reservoir classification criteria were formulated, categorizing reservoirs into four grades. Effective reservoirs in deep layers were mainly grades Ⅱ and Ⅲ. The distribution of effective reservoirs in the deep layers was evaluated across key stratigraphic intervals, revealing the graded distribution of reservoirs in deep zones of the first and third member of Funing Formation, the third submember in the first member of Dainan Formation in Gaoyou Sag, and the second member of Funing Formation in Jinhu Sag. The favorable areas of effective reservoirs in the deep layers of each stratigraphic system in each Sag were finally determined. (2) Through the analysis of deep oil and gas enrichment mechanisms, and according to the dynamic conditions of oil and gas injection, models for calculating reservoir potential energy, fluid potential, and source-reservoir pressure differences were established. Subsequently, a model for calculating the reservoir injection potential energy index were established based on the above models. Finally, the obtained reservoir injection potential energy index was used to assess the probability of oil and gas accumulation, providing technical support for the selection of favorable oil and gas accumulation zones in deep layers. (3) Subaqueous distributary channels and beach-bar sand bodies were effective reservoirs for deep oil and gass. To address the challenge of effective reservoir prediction in thin sandstone-mudstone interbeds within favorable oil and gas accumulation zones in selected deep layers, an integrated technical suite for effective reservoir prediction was developed. This technique, tailored to different sand body types such as channels and beach bars, integrated pre-stack and post-stack multi-attribute analysis. It leveraged geological, petrophysical, seismic, statistical, and other disciplinary theories to provide a comprehensive approach to reservoir prediction. Based on the distinction between sandstone and mudstone, this suite included six techniques for reservoir prediction: effective reservoir modeling based on petrophysical analysis, post-stack multi-parameter inversion constraint method, pre-stack and post-stack joint inversion method, seismic attribute threshold analysis method, seismic multi-attribute neural network prediction method, and SP curve reconstruction for acoustic curve. These techniques collectively improved the prediction accuracy of effective reservoirs in deep layers. These research findings provide theoretical guidance and technical support for the expansion of deep oil and gas exploration. Significant exploration progress has been made in deep layers such as slope zones, fault zones, and deep sag zones, enabling the expansion of deep oil and gas exploration. In the future, the research directions for addressing challenges in deep oil and gas exploration are clarified, which are continuing to consolidate and expand deep exploration to support the increase in oilfield reserves and production.

Yuta Watanabe

In this paper, by refining approximation theorems for holomorphic sections of adjoint line bundles, it is proved that the regular locus of a weakly pseudoconvex complex space admitting a positive line bundle can be holomorphically embedded into a complex projective space. As an application of approximation theorems, it is shown that the Union problem can be solved for weakly pseudoconvex manifolds.

Zhengyang Yu, Akio Hayakawa, Masato Ishii et al.

Autoregressive video diffusion models (AR-VDMs) show strong promise as scalable alternatives to bidirectional VDMs, enabling real-time and interactive applications. Yet there remains room for improvement in their sample fidelity. A promising solution is inference-time alignment, which optimizes the noise space to improve sample fidelity without updating model parameters. Yet, optimization- or search-based methods are computationally impractical for AR-VDMs. Recent text-to-image (T2I) works address this via feedforward noise refiners that modulate sampled noises in a single forward pass. Can such noise refiners be extended to AR-VDMs? We identify the failure of naively extending T2I noise refiners to AR-VDMs and propose AutoRefiner-a noise refiner tailored for AR-VDMs, with two key designs: pathwise noise refinement and a reflective KV-cache. Experiments demonstrate that AutoRefiner serves as an efficient plug-in for AR-VDMs, effectively enhancing sample fidelity by refining noise along stochastic denoising paths.

Yuan Feng, Li Zhou

Refinement is a fundamental technique in the verification and systematic development of computer programs. It supports a disciplined approach to software construction through stepwise refinement, whereby an abstract specification is gradually transformed into a concrete implementation that satisfies the desired requirements. Central to this methodology is the notion of a refinement order, which guarantees that each refinement step preserves program correctness. This paper presents the first comprehensive study of refinement orders for quantum programs, covering both deterministic and nondeterministic settings under total and partial correctness criteria. We investigate three natural classes of quantum predicates: projectors, representing qualitative properties; effects, capturing quantitative properties; and sets of effects, modeling demonic nondeterminism. For deterministic quantum programs, we show that refinement with respect to effect-based and set-of-effects based specifications coincides with the standard complete-positivity order on superoperators, whereas refinement induced by projector-based specifications can be characterized by the linear span of Kraus operators. For nondeterministic quantum programs with set-of-effects based specifications, we establish precise correspondences with classical domain-theoretic notions: the Smyth order characterizes refinement under total correctness, while the Hoare order characterizes refinement under partial correctness. Moreover, effect-based and projector-based specifications lead to strictly weaker refinement orders.

Oliver Knill

The soft Barycentric refinement preserves manifolds with or without boundary. In every dimension larger than one, there is a universal spectral central limiting measure that has affinities with the Barycentric limiting measure one dimension lower. Ricci type quantities like the length of the dual sphere of co-dimension-2 simplex stay invariant under soft refinements. We prove that the dual graphs of any manifold can be colored with 3 colors, which is in the 2-dimensional case a special case of the Groetzsch theorem. It follows that the vertices of a soft Barycentric refined q-manifold G' can be colored by q+1 or q+2 colors.

Seyed Reza ASADOLAHPOUR, Zeyun JIANG, Helen LEWIS et al.

In order to predict phase distributions within complex pore structures during two-phase capillary-dominated drainage, we select subsamples from computerized tomography (CT) images of rocks and simulated porous media, and develop a pore morphology-based simulator (PMS) to create a diverse dataset of phase distributions. With pixel size, interfacial tension, contact angle, and pressure as input parameters, convolutional neural network (CNN), recurrent neural network (RNN) and vision transformer (ViT) are transformed, trained and evaluated to select the optimal model for predicting phase distribution. It is found that commonly used CNN and RNN have deficiencies in capturing phase connectivity. Subsequently, we develop a higher-dimensional vision transformer (HD-ViT) that drains pores solely based on their size, regardless of their spatial location, with phase connectivity enforced as a post-processing step. This approach enables inference for images of varying sizes and resolutions with inlet-outlet setup at any coordinate directions. We demonstrate that HD-ViT maintains its effectiveness, accuracy and speed advantage on larger sandstone and carbonate images, compared with the microfluidic-based displacement experiment. In the end, we train and validate a 3D version of the model.

Li Jiayi, Liu Xiuquan, Xu Liangbin et al.

The traditional evaluation on wave induced fatigue of subsea wellhead does not consider the changes in the composition of subsea wellhead structure at different drilling stages,resulting in limited accuracy in wave induced fatigue evaluation of subsea wellhead. A fine evaluation method for wave induced fatigue of subsea wellhead in full drilling process based on global-local interaction analysis was proposed. A local fine model for subsea wellhead with complex structure and a global model for platform-riser-subsea wellhead-conductor coupling system in the full drilling process were built. The interaction mechanism between global model and local model was studied. The transitive relation between global model and local stress was constructed to finely evaluate the wave induced fatigue damage of subsea wellhead in full drilling process. The results show that the changes in the composition of subsea wellhead structure have almost no effect on the wave induced fatigue of the riser system and the conductor far away from the mud line in full drilling process,but have a significant impact on the wave induced fatigue of local part of subsea wellhead and the conductor near the mud line. At the second stage of drilling,the wave induced fatigue damage to the subsea wellhead is the most severe,and the bottom of the subsea wellhead is always at the fatigue limit position. The conclusions are helpful in judging the fatigue of subsea wellhead during drilling and ensuring the safety of drilling operations.

Mahdi Hosseini, Seyed Hayan Zaheri, Ali Roosta

Abstract During the development of a field, many fluid samples are taken from wells. Selecting a robust fluid sample as the reservoir representative helps to have a better field characterization, reliable reservoir simulation, valid production forecast, efficient well placement and finally achieving optimized ultimate recovery. First, this paper aims to detect and separate the samples that have been collected under poor conditions or analyzed in a non-standard way. Moreover, it introduces a novel ranking method to score the samples based on the amount of coordination with other fluid samples in the region. The dataset includes 136 fluid samples from five reservoirs in Iranian fields, each of them consisting of 21 key parameters. Five acknowledged machine learning based anomaly detection techniques are implemented to compare fluid samples and detect those whose results deviate from others, indicating non-standard samples. To ensure the proper detection of outlier data, the results are compared with the traditional validation method of gas-oil ratio estimation. All five outlier detection methods demonstrate acceptable performance with average accuracy of 79% compared to traditional validation. Furthermore, the fluid samples with the highest scores in scoring-based algorithms are introduced as the best reservoir’s representative fluid. Finally, fuzzy logic is used to obtain a final score for each sample, taking the results of the six methods as input and ranking the samples based on their output score. The study confirms the robustness of the novel approach for fluid validation using outlier detection techniques and the value of machine learning and fuzzy logic for sample ranking, excelling in considering all critical fluid parameters simultaneously over traditional methods.

A. Bahri, M. Bendersky, F. R. Cohen et al.

We give a geometric method for determining the cohomology groups of a polyhedral product under suitable freeness conditions or with coefficients taken in a field. This is done by considering first the special case for which the pairs of spaces are wedge decomposable. We derive a decomposition for these polyhedral products which resembles a Cartan formula. The theory of symmetric products is used then to generalize the result to polyhedral products involving arbitrary pairs. This leads to a direct computation of the Hilbert-Poincaré series and to other applications.

Elena A. Lebedeva

In the paper we design a Parseval wavelet frame with a compact support. The corresponding refinement mask uniformly approximates an arbitrary continuous periodic function $f$, $f(0)=1$, $|f(x)|^2+|f(x+π)|^2\le 1$. The refinable function has stable integer shifts.

Zhu Junlong, Li Qingping, Qin Rui et al.

As a key component of subsea Christmas tree,subsea chokes work in deep water,high pressure and low temperature environment.It is easy to produce natural gas hydrates to block chokes and downstream pipelines due to J-T effect.Therefore,it is very important to study the temperature range downstream of the chokes and compare the hydrate temperature-pressure curves for production safety.In this paper,based on the theories of mass,kinetic energy,energy conservation,heat transfer and fluid mechanics,taking the Cameron subsea chokes as the study object,the subsea choke throttling model was established with natural gas condensate as the medium to study the throttling temperature drop of natural gas through chokes.Fluent software was used to obtain the flow field distribution and temperature distribution of subsea chokes from single-phase methane to gas-liquid two-phase flow.Finally,by using the established model,the effects of inlet pressure,outlet pressure,inlet temperature,gas-liquid ratio and choke opening on the throttling temperature drop and choke flow field distribution were studied.The results show that when fluid passes through the sudden contraction section of subsea chokes,the downstream temperature drops sharply due to J-T effect,and when the temperature drops to a sufficiently low level,hydrate will precipitate and block the pipeline.Controlling the pressure and temperature above the hydrate formation point in production could avoid ice blockage.In addition,the choke opening will indirectly affect the pressure difference,so too small opening is not applicable in operations.The results provide guidance for safe production in deepwater oil and gas fields.

LI Hongwei, YUAN Jian, ZHAO Zhichuan et al.

The J<i>S</i>₂² gas reservoir in the Xinchang Gas Field is the microfacies of distributary channel deposit in the delta front subfacies. The mudstone interlayer developed in some sandstone layers form the channel sediments twice. For some gas wells with double developed sand layers, the perforation and fracturing are carried out only in one of the two layers, while the rest may be blocked by mudstone interlayer, resulting in its untapped situation. However, the understanding of the distribution rule of the mudstone interlayer is unclear, and the sealing performance of interlayers hasn't been verified by precedent, limiting the development of the remaining gas in this reservoir. By the establishment of the logging curve identification standards of the interlayer, the interlayer of every single well is identified and divided, and the development scale and plane distribution rule of the interlayer are studied. And then, the old wells are used to explore the vertical sealing performance of the mudstone interlayer inside the channel sandstone. The study results show that the mudstone interlayer, with the thickness of 1~15 meters, is mainly distributed in the east of the J<i>S</i>₂² gas reservoir, and has the characteristic of one thin-line area laying between two thick areas along the distribution direction. In the application of potential tapping of Well-A1, the production obviously increase after the perforation and sand-fracture, firstly verifying that the mudstone interlayer with the thickness of 10 meters in A1 well area has better vertical sealing performance. This achievement will provide new ideas and practical basis for the development of remaining gas in the J<i>S</i>₂² gas reservoir, and is of great significance for improving the recovery rate of the entire gas reservoir.

Afanas'eva M.A., Chzhan Ee

The deltas developed widely in several faulted basins in China. As a kind of important oil and gas reservoir, it is very necessary to make clear the internal architecture for prediction of the residual oil distribution in the late period of oilfield development．Therefore taking the 5th oil productive level of Kongdian Formation in the faulted block 195 of Y oil field as an example the reservoir architecture of delta front was characterized quantitatively based on core data well logs and production data．The main reservoir architecture elements off an delta front consists of distributary channel, estuary sandbar, sheet sand and inter channel．The division scheme was established for the division of reservoir architecture of delta front into different levels and the markers was defined for the identification of distributary channel estuary sandbar and inner accretion body in estuary sandbar．Taking single sandbody as a unit data being acquired from field outcrop defined as delta in the Xiguayuan Formation of Luanping Basin were matched to derive a formula for quantitative characterization of distributary channel and estuary sandbar, calculation of the dip of interbed innerestuary sandbar and prediction of the development of inner accretion body．The residual oil controlled by a single genetic sand body distributes at the top of the margin of sandbars and channels and the distribution of residual oil controlled by the interlayer should be analyzed according to the specific rhythm of the sandbar.

Yulia Petrova, Bradley J. Plohr, Dan Marchesin

We examine the vanishing adsorption limit of solutions of Riemann problems for the Glimm-Isaacson model of chemical flooding of a petroleum reservoir. A contact discontinuity is deemed admissible if it is the limit of traveling waves or rarefaction waves for an augmented system that accounts for weak chemical adsorption onto the rock. We prove that this criterion justifies the admissibility criteria adopted previously by Keyfitz-Kranzer, Isaacson-Temple, and de Souza-Marchesin, provided that the fractional flow function depends monotonically on chemical concentration. We also demonstrate that the adsorption criterion selects the undercompressive contact discontinuities required to solve the general Riemann problem in an example model with non-monotone dependence.

Michael Borkowski, Niki Vazou, Ranjit Jhala

Practical checkers based on refinement types use the combination of implicit semantic sub-typing and parametric polymorphism to simplify the specification and automate the verification of sophisticated properties of programs. However, a formal meta-theoretic accounting of the soundness of refinement type systems using this combination has proved elusive. We present λ_RF a core refinement calculus that combines semantic sub-typing and parametric polymorphism. We develop a meta-theory for this calculus and prove soundness of the type system. Finally, we give a full mechanization of our meta-theory using the refinement-type based LiquidHaskell as a proof checker, showing how refinements can be used for mechanization.

David Wadden, Nikita Gupta, Kenton Lee et al.

We introduce the task of entity-centric query refinement. Given an input query whose answer is a (potentially large) collection of entities, the task output is a small set of query refinements meant to assist the user in efficient domain exploration and entity discovery. We propose a method to create a training dataset for this task. For a given input query, we use an existing knowledge base taxonomy as a source of candidate query refinements, and choose a final set of refinements from among these candidates using a search procedure designed to partition the set of entities answering the input query. We demonstrate that our approach identifies refinement sets which human annotators judge to be interesting, comprehensive, and non-redundant. In addition, we find that a text generation model trained on our newly-constructed dataset is able to offer refinements for novel queries not covered by an existing taxonomy. Our code and data are available at https://github.com/google-research/language/tree/master/language/qresp.

Qiang Jie, Qi Yuekui, Liu Xueguang et al.

At present, the casing failure injection wells in Dagang Oilfield account for 9.2% of the total number of oil and water wells in the oilfield, which affects the stable production and production increase of the oilfield. In view of this, the technological plan of using expansion pipe patch technology to repair the casing failure well was established. After having briefly introduced the principle of expansion pipe patch technology, the structural composition of patch tool string, the operation technology and the technological characteristics, this article carried out mechanical property test on the expansion pipe patch tubular goods, optimized high strength and low cost tubular goods 20G, used the multi-stage force-increasing cold drawing expansion forming process of &#248;139.7 mm casing to form large drift diameter expansion pipe patch technology, and subsequently carried out expansion pipe patch laboratory test. The test results show that the expansion pressure of the expansion pipe is 14～18 MPa, and the expansion rate of it is greater than 8.94%; the expansion pipe and the casing can withstand 500 kN tension without relative movement; the anchoring seal between the tubular goods and the casing is reliable, and the large drift diameter of the patch section is realized. This technology has been successfully applied in Well Jia47-7 of Dagang Oilfield, and the expansion pipe patch achieves the expected effect of plugging the leakage point of the pipe, which provides technical support and reference for the repair of casing failure injection wells in the oilfield.

Lichun KUANG, He LIU, Yili REN et al.

Aiming at the actual demands of petroleum exploration and development, this paper describes the research progress and application of artificial intelligence (AI) in petroleum exploration and development, and discusses the applications and development directions of AI in the future. Machine learning has been preliminarily applied in lithology identification, logging curve reconstruction, reservoir parameter estimation, and other logging processing and interpretation, exhibiting great potential. Computer vision is effective in picking of seismic first breaks, fault identification, and other seismic processing and interpretation. Deep learning and optimization technology have been applied to reservoir engineering, and realized the real-time optimization of waterflooding development and prediction of oil and gas production. The application of data mining in drilling, completion, and surface facility engineering etc. has resulted in intelligent equipment and integrated software. The potential development directions of artificial intelligence in petroleum exploration and development are intelligent production equipment, automatic processing and interpretation, and professional software platform. The highlights of development will be digital basins, fast intelligent imaging logging tools, intelligent seismic nodal acquisition systems, intelligent rotary-steering drilling, intelligent fracturing technology and equipment, real-time monitoring and control of zonal injection and production.