Hanpeng Cai

Dr. Hanpeng Cai

orcid

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
18 October 2023 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
XHXiangyu Hu
Xiangyu Hu
HCHanpeng Cai
Hanpeng Cai
MAMamoun Alazab
Mamoun Alazab
WZWei Zhou
Wei Zhou
MHMohammad Sayad Haghighi
Mohammad Sayad Haghighi
VWVincent Sheng Wen
Vincent Sheng Wen
Applying artificial intelligence (AI) to data from Industrial Internet of Things (IIoT) devices is a novel direction in geological studies. However, privacy and security concerns hinder the sharing of data, thus affecting the performance of current AI-based approaches. In this article, we propose a novel data management style to address the privacy and security issues in joint hydrocarbon explorations. Federated learning can facilitate the analysis of multiple datasets without the need to share them, protecting private information of different companies in a virtual joint venture. We use the inference of petroleum reservoirs in karst stratigraphy as a case study. A federated learning-based enterprise data management framework is proposed to virtually integrate the information from different organizations. Our key contributions are summarized as follows. 1) A method for karst identification and inference is proposed, which uses neural networks to recognize the size of petroleum reservoirs in different karst areas. 2) A federated learning algorithm is applied to virtually aggregate data samples from different companies. 3) The performance of the new privacy-preserving integration model is compared with those of the individual/local deep learning models. Our results show that the proposed approach can substantially improve the accuracy of petroleum reservoir explorations.

1 August 2019 in Interpretation
Verified
Journal Article
Closed access
BSBin She
Bin She
YWyaojun wang
yaojun wang
ZLZhining Liu
Zhining Liu
HCHanpeng Cai
Hanpeng Cai
WLWei Liu
Wei Liu
GHGuangmin Hu
Guangmin Hu
We have addressed the seismic impedance inversion problem, which is often ill posed because of inaccurate and insufficient data. The approach taken is based on dictionary learning and sparse representation. By shifting a patch window of fixed size on all well-log data, a large number of small overlapping patches are generated. Then regarding these patches as a training set and using K-singular value decomposition algorithm, we obtain a dictionary that describes the common features of subsurface models within the current survey area. On the basis of the assumption that the subsurface geology has similarity and lateral continuity to some extent, the dictionary is used to approximate each model via sparse representations over the learned dictionary. In particular, we impose the sparse representations as additional constraints into the inversion procedure, leading to a well-defined objective function that can not only fit the observed seismic data but also honor the features of the well-log data. We adopt a coordinate descent strategy to solve this objective function. Meanwhile, to enforce lateral continuity in the inverted models, we use an additional stage in which we use the nonlocal similarity information that is extracted from seismic data as spatial coherent prior knowledge to refine the estimated models. Compared with several traditional impedance inversion methods, our algorithm can produce solutions of much higher quality qualitatively and quantitatively.

1 September 2020 in Geophysical Prospecting
Verified
Journal Article
Closed access
PDF file
KLKunhong Li
Kunhong Li
ZLZhining Liu
Zhining Liu
BSBin She
Bin She
HCHanpeng Cai
Hanpeng Cai
YWyaojun wang
yaojun wang
GHGuangmin Hu
Guangmin Hu

24 May 2022 in Geophysics
Verified
Journal Article
Closed access
TCTing Chen
Ting Chen
JLJiandong Liang
Jiandong Liang
PLPengfei Li
Pengfei Li
HCHanpeng Cai
Hanpeng Cai
GHGuangmin Hu
Guangmin Hu
YWyaojun wang
yaojun wang
Geostatistical inversion is a technique widely used in seismic inversion. It builds the posterior distribution by incorporating prior information. Abundant prior information for geostatistical inversion is provided by the spatial variability of well log represented by variograms. However, a variogram can only describe the spatial variability of lithologic parameters at a single point. Describing the structural relationship of rock deposition in the vertical direction (the sedimentary rhythm) requires a continuous curve, rather than one single point. We have developed a geostatistical seismic inversion method constrained by sedimentary structural features to enhance the consistency between inversion results with the sedimentary features of a reservoir. This method first uses sparse representation to learn the sedimentary structural features of a reservoir. Then, the structural features are taken as prior information of geostatistical inversion to improve the accuracy of inversion results. The novelty of our method is that the element describing the spatial variability of model parameters is extended from a single point to a sedimentary structural block. We derive an analytical expression of posterior distribution of model parameters with constraints of sedimentary structural features and realize a data-driven geostatistical seismic inversion. Experiments on synthetic and field data indicate promising results consistent with the sedimentary features of reservoir.

5 December 2022 in Geophysics
Verified
Journal Article
Closed access
TCTing Chen
Ting Chen
YWyaojun wang
yaojun wang
KLKunhong Li
Kunhong Li
HCHanpeng Cai
Hanpeng Cai
GYGang Yu
Gang Yu
GHGuangmin Hu
Guangmin Hu
The inversion of thin-bed reservoirs using seismic data is important for precise reservoir prediction. Most of the current thin-bed reservoir inversion methods are constrained by facies or sedimentary characteristics in reservoirs. The accuracy of the sedimentary characteristics severely affects the precision of the inversion results. There are differences in sedimentary patterns among different facies, but the same facies have similar sedimentary characteristics, and the similarity brings the sparsity of sedimentary structure within a facies. Thus, various representations should be used to describe the differences in the sedimentary patterns between facies. After analyzing the sparse representation (SR) patterns for different facies, a prestack seismic inversion method constrained by facies-controlled sedimentary structural features is developed. Our method first uses seismic attributes that provide rich details of seismic features to divide facies. Then, the sedimentary structural features within each facies are extracted. The inversion results are represented by structural features in the corresponding facies. We have used the represented results as the initial value of the inversion to iterate. Using this iterative inversion framework, we obtain results that are consistent with the sedimentary features of each facies. The representation of reservoir parameters by different SR patterns can avoid the mixing of sedimentary structural features in different facies. Synthetic and field examples are used to verify the effectiveness of our method. The method obtains inversion results that are consistent with seismic data and the distribution of thin-bed in well logs. Therefore, the method can be successfully applied to thin-bed sandstone reservoirs.

27 May 2022 in Geophysics
Verified
Journal Article
Closed access
CZCheng Zhou
Cheng Zhou
RZRuoshui Zhou
Ruoshui Zhou
XZXianglin Zhan
Xianglin Zhan
HCHanpeng Cai
Hanpeng Cai
XYXingmiao Yao
Xingmiao Yao
GHGuangmin Hu
Guangmin Hu
Fault surface extraction plays a vital role in structural interpretation and structural modeling, which can enhance our understanding of geologic structures. Significant effort has been invested in fault surface extraction in the past few years. These contributions can be roughly divided into two types: one constructs fault sticks or fault patches and links them to form fault surfaces, and the other constructs a complete fault surface by roughly determining the location of a fault. The former is likely to extract incomplete fault surfaces in areas with low data quality or complex fault structures. The latter has problems dealing with faults in the case of complex fault structures, such as intersecting faults. To effectively address these problems, we have developed a novel fault extraction workflow from a global perspective, which can extract fault surfaces on the premise of obtaining the fault distribution of the entire data set. First, we introduce a multilayer complex network to characterize the overall distribution of faults, in which nodes represent faults, node attributes cover information, such as the location of the faults, and edge attributes give the relationship between faults. Then, for each fault represented by the node, we develop a fault surface extraction method based on computational topology, which can guarantee the completeness of the fault surfaces. We apply our method to synthetic data and field seismic data to evaluate the method quantitatively and qualitatively, respectively. The results indicate that the proposed approach extracts the fault effectively, ensures the completeness of the fault surface, and has high performance in areas with complex faults. Thus, the approach can work well in complex situations, such as intersecting faults.