- Hydrocarbon exploration and reservoir analysis
- Geological and Geophysical Studies
- Geological Studies and Exploration
- Geoscience and Mining Technology
- Geochemistry and Geologic Mapping
- Coal Properties and Utilization
- Nonlinear Dynamics and Pattern Formation
- Geological formations and processes
- Petroleum Processing and Analysis
- Geochemistry and Geochronology of Asian Mineral Deposits
- Chaos control and synchronization
- Geology and Paleoclimatology Research
- Drilling and Well Engineering
- Generative Adversarial Networks and Image Synthesis
- Physics and Engineering Research Articles
- Mechanical and Thermal Properties Analysis
- Reinforcement Learning in Robotics
- Paleontology and Stratigraphy of Fossils
- Soil, Finite Element Methods
- Neural Networks and Applications
- Geomechanics and Mining Engineering
- Coal and Its By-products
- Human Pose and Action Recognition
- Simulation Techniques and Applications
- Target Tracking and Data Fusion in Sensor Networks
Microsoft Research Asia (China)
2022-2024
Shandong University of Science and Technology
2006-2023
Sinopec (China)
2021
State Key Laboratory of Continental Dynamics
2014
Chinese Academy of Geological Sciences
2011-2014
Czech Academy of Sciences, Institute of Geology
2014
Research Institute of Petroleum Exploration and Development
2011
China University of Geosciences (Beijing)
2009
SI2 Technologies (United States)
2003
In a recent paper [2002 Phys. Rev. Lett. 88 174102], Bandt and Pompe propose permutation entropy (PE) as natural complexity measure for arbitrary time series which may be stationary or nonstationary, deterministic stochastic. Their method is based on comparison of neighbouring values. This further develops PE, proposes the concept fine-grained PE (FGPE) defined by order pattern magnitude difference between excludes case where vectors with distinct appearance are mistakenly mapped onto same...
In this report, the influence of pre-oxidation degree and ventilation flow on parameters spontaneous combustion coal (temperature, gas concentration, exothermic intensity) was studied in six sets programed temperature experiments. The experimental results showed that exerted a positive effect early stage coal-oxygen recombination reaction, but an inhibitory later oxidation reaction. Air supply rate had correlation with initial samples 90 °C parameters. negative 140 samples. Meanwhile,...
After coal seam mining, the residual is soaked with accumulated water in goaf, and its spontaneous combustion characteristics were changed after air-dried. To study reoxidation of air-dried coal, temperature-programmed experiments carried out, cross point temperatures index gases investigated. Results showed that temperature raw (146.3 °C) was reduced to 137.1 °C it pre-oxidized at 90 °C. The water-soaked, (96 h) 122.5 °C, while (90 111.5 Although CO produced initial slow oxidation phase,...
Analysis of reservoir sensitivity to velocity, water, salt, acid, alkali and stress is critical for protection. To study the tight sandstone at different formation depths (effective stress) water conditions (pH, salinity, fluid velocity), a series dynamic core flow tests under pH, effective were performed on samples from reservoirs Upper Triassic Yanchang 8 (T3y8) Member conventional Middle-Lower Jurassic Yan'an 9 (J1-2y9) in Ordos Basin. The results indicate that, compared with reservoirs,...
The low-temperature oxidation law of coal and rock mass is the basis to study spontaneous combustion in goafs. In this paper, laws coal, oil shale, mixtures shale were studied by using laboratory programmed heating experiments combined with a field beam tube monitoring system. results from showed that heat released was less than coal. Coal had lower carbon monoxide (CO)-producing temperature mixture obvious inhibiting effects on CO production an average concentration about 38% for Index...
A substantial amount of petroleum was recently discovered in the Carboniferous volcanic reservoir Chepaizi Uplift western Junggar Basin, yet source is still indefinitive. Geochemical investigation indicates that oils from eastern and are characterized by different facies, although they all typically lacustrine origin. The exhibit a restricted, clastic starved, highly reducing hypersaline condition during rock deposition, which distinct oils. were subjected to biodegradation ranging rank 6 9,...
Abstract Offline policy evaluation, evaluating and selecting complex policies for decision‐making by only using offline datasets is important in reinforcement learning. At present, the model‐based evaluation (MBOPE) widely welcomed because of its easy to implement good performance. MBOPE directly approximates unknown value a given Monte Carlo method estimated transition reward functions environment. Usually, multiple models are trained, then one them selected be used. However, challenge...
Human body pose transfer is to transform the character image from source target pose. In recent years, research has achieved great success in transforming human image, but it still insufficient detailed texture of generated image. To solve above problems, a new two-stage TPIT network model proposed process pose-generated The first stage self-learning module, which extracts features by learning itself and further improves appearance details other change figure gradually Then, feature...
The Junggar Basin is one of the largest and most petroliferous superimposed petroleum basins in China. This study focussed on characteristics unconformity between Jurassic Cretaceous (J/K) its influence hydrocarbon accumulation Yongjin area, Basin. Based seismic reflection wireline well logging response, we identified structure unconformity, which consists three layers: basal conglomerate layer, weathered clay semi‐leached layer. Geochemical studies show that crude oils Lower Qingshuihe (K 1...
For the Multi-Beam Echo Sounder (MBES) data denoising problem, noise is divided into near-ground and outlier away from seabed. The algorithm in this paper considers spatial distribution characteristics of removes two kinds by initial precise denoising. first uses octree index to organize point cloud noise. Then combined with Bayesian estimation theory statistical methods remove This designs experiments process MBES compare it filter PCL library. experimental results show that compared...
Learning dynamics governed by differential equations is crucial for predicting and controlling the systems in science engineering. Neural Ordinary Differential Equation (NODE), a deep learning model integrated with equations, popular recently due to its robustness irregular samplesand flexibility high-dimensional input. However, training of NODE sensitive precision numerical solver, which makes convergence unstable, especially ill-conditioned dynamical systems. In this paper, reduce reliance...
Based on piecewise monotone interval maps and linear coupling, we study order patterns of spatiotemporal chaos. The forbidden are found to arise mainly from the reduction curve intersections due time invariance chaotic maps. It is proved that couplings may destroy invariance, create conditions for increasing intersections. We analyze effects map, coupling strength number patterns. Simulation results illustrative examples all confirm correctness theoretical results.
Abstract The Upper Permian around Bogda Mountain develops favorable reservoir sand bodies which show good oil and gas display. Further study of the sedimentary characteristics evolution models has guiding significance towards petroleum exploration. Through detailed obervation measurement in typical profile periphery Mountain, combined with well drilling logging data core Chaiwobao depression, strata characteristics, facies types distribution features studying area were carefully analysized,...