Xiaobo Peng

Xiaobo Peng

Auto-generated profile, based on publicly available information

Xiaobo Peng

School of Geophysics and Petroleum Resources, Yangtze University, Wuhan 430100, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 February 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
CWCunwu Wang
Cunwu Wang
XPXiaobo Peng
Xiaobo Peng
GHGang Han
Gang Han
YZYan Zhao
Yan Zhao
YZYihao Zhu
Yihao Zhu
MZMing Zhao
Ming Zhao
This paper proposes a coal structure prediction technology based on deep learning, which uses logging data to achieve single-well prediction of the coal structure. This paper introduces the genetic algorithm (GA) to optimize the BP neural network, which can speed up its convergence to the global optimal solution, improve its training speed, and avoid the problems of easily producing the local optimal value and requiring a long training time. Taking the main coal seam of the Shizhuang block in the south of the Qinshui Basin as the research object and using the coal core data and logging data of nine parameter wells, the mapping relationship between the logging curve and coal structure is constructed based on the GA-BP neural network structure, and the coal structure is predicted. The prediction results are highly consistent with the coal structure measured from coal core sampling, with only a small error, and the prediction accuracy is 90%. It is shown that the GA-BP neural network structure can be used to effectively identify the coal structure, as well as predict the coal structure of uncored wells. Moreover, the findings of this study will be helpful for efforts to study the distribution law of the coal structure.

22 February 2020 in Computational Intelligence
Verified
Journal Article
Closed access
PDF file
XPXiaobo Peng
Xiaobo Peng
YCYanfen Chang
Yanfen Chang

10 August 2021 in Carbonates and Evaporites
Verified
Journal Article
Closed access
PDF file
XPXiaobo Peng
Xiaobo Peng
AAAseel M. Aljeboree
Aseel M. Aljeboree
ATAnton Timoshin
Anton Timoshin
SNSeyed Mohammad Mehdi Nassabeh
Seyed Mohammad Mehdi Nassabeh
ADAfshin Davarpanah
Afshin Davarpanah

24 June 2019
Verified
Conference Paper
Open access
TYTianyun Yuan
Tianyun Yuan
XPXiaobo Peng
Xiaobo Peng
DZDongdong Zhang
Dongdong Zhang
LLLin Li
Lin Li
There is a demanding need for rapid replication or reproduction of physical objects. In many industries such as turbine manufacturing, aerospace, biomedicine, culture heritage etc., the CAD files of parts of interest are usually unavailable or inaccessible. In some situations, such as automotive styling, the designed parts are created by the designer using clay, wood, or foam rubber. The conventional method of reproducing parts includes two main processes, i.e., Reverse Engineering (RE) and manufacturing. The conventional procedure of reproducing a part includes several steps. First, the cloud point data is collected from the surface of an existing part by 3D scanning. Next, the CAD model of the object is reconstructed using professional software. Afterward, the CAD model is converted into a facet model. The facet model is then sliced into layers by the slicing software. A G-code file is generated and imported to the 3D printer for final manufacturing. The whole process is far from being automatic. The steps result in rather expensive remodeling computations, large file transmissions, and highly accumulated approximation errors. Laborious work and professional knowledge are required for all steps. Users have to be familiar with various modeling software, which usually requires years of training for them to be efficient and productive.

8 July 2026 in Processes
Verified
Journal Article
Open access
PDF file
XZXiao Zhang
Xiao Zhang
QCQi Chang
Qi Chang
ZWZhen Wang
Zhen Wang
XPXiaobo Peng
Xiaobo Peng
SZShi Jie Zhu
Shi Jie Zhu
The fracture–cave carbonate reservoir in Unit S91 of the Tahe Oilfield is jointly controlled by strike-slip fault activity, karstification, and later-stage fracture development, resulting in reservoir spaces characterized by strong heterogeneity, strong discreteness, and multi-scale superimposition. The inter-well connectivity of this type of reservoir is not governed by the size of a single fracture–cave body or local fracture density, but rather by the spatial configuration among the main controlling fault, the associated fracture network, and the fracture–cave reservoir bodies. As the reservoir enters the middle–high-water-cut development stage, the production differential between dominant connecting channels and weakly connected fracture–cave bodies further enlarges, leading to marked heterogeneity in the remaining oil distribution. Integrating post-stack seismic data, fracture prediction, RGB attribute fusion, production performance, and numerical simulation data, this paper constructs a main fault-controlled fracture–vug coupling probability (MFC-FVCP) model under the constraint of the main controlling fault. Unlike conventional multi-attribute fusion methods that mainly enhance seismic anomaly visualization, the MFC-FVCP model transforms the main fault constraint, fracture connectivity, and fracture–cave reservoir-body effectiveness into a unified coupling probability. The model uses three core components—the main fault response field, the fracture attribute response field, and the fracture–cave reservoir body response field—to characterize the fault-control effect, fracture-network continuity, and effective reservoir-body response, respectively. By evaluating the coupling probability, the inter-well connectivity potential is assessed, the dominant connectivity areas where fractures and fracture–cave bodies synergistically develop under the constraint of the main controlling fault are identified, and potential remaining oil targets are clarified. The predicted connectivity pattern was further constrained by production performance, nitrogen injection response, and staged oil saturation simulation, which improves the reliability of remaining oil enrichment prediction. The results show that the T74 layer is the dominant development interval of fracture–cave reservoir bodies in Unit S91. These fracture–cave bodies are mainly distributed along the main controlling fault and associated fracture zones in beaded, chain-like, and banded patterns, exhibiting distinct fault-and-fracture control characteristics. Potential point A near well TK858XCH features both good reservoir physical properties and insufficient sweep efficiency, making it a key target for subsequent injection–production adjustment and remaining oil tapping. The MFC-FVCP model can incorporate static seismic responses, fracture–cave spatial structures, and dynamic development responses into a unified evaluation framework, providing a quantitative basis for characterizing inter-well connectivity and identifying remaining oil enrichment areas in fracture–cave carbonate reservoirs.