1. Insititute of Gravity and Magnetic Technology,Chang’an University,Xi’an 710054,China 2. College of Geology Engineering and Geomatics,Chang’an University,Xi’an 710054,China 3. Key Laboratory of Western China’s Mineral Resources and Geological Engineering,Ministry of Education,Chang’an University,Xi’an 710054,China 4. China Aero Geophysivey and Remote Sensing Center for Natural and Resources,Beijing 100083,China
Accurately characterizing rock types and their structural relationships can provide important information for researches on energy and mineral exploration, deep structure and tectonic, and so on. At present, geophysical data can be used to identify lithology through the differences between the corresponding physical property parameters of different rocks (such as density, susceptibility, resistivity, speed,etc.). However, different rock physical properties often coincide at a certain degree, on the other hand the result of lithological identification is not accurate enough when a single physical property is used. Therefore, it is of great significance to use multi-source data for lithological identification. Bayesian method belongs to the statistical classification methods, which relies on probability for classification, and the calculation of probability density depends on the correlation between sample attributes. On such a basis, the authors introduce the Bayesian probability model based on adaptive kernel density estimation to lithology identification. This method has good adaptability for many different types of physical property parameters and such the predicted lithology classification results with probability parameters, fuzzy intervals, and a variety of lithology classification results. This method has a strong scalability that can process both parametric and non-parametric information at the same time to maximally the known geological information and physical parameters. Synthetic models prove that this method has the ability to provide more stable and accurate results of lithology recognition compared with the methods of traditional Gaussian algorithm and fixed bandwidth kernel density estimation.
Jin J, Liu L J, Shao Y, et al. Discussion on identifying method for identification of lithologic traps in Junggar Basin by comprehensive geophysical method[J]. Oil Geophysical Prospecting, 2002,37(3):287-290, 299.
Hong Z, Zhang M G, Su M J. The applicability and limitations of the seismic wave-form classification technology to the identification of lithological facies[J]. Geophysical and Geochemical Exploration, 2013,37(5):904-910.
[5]
Tjelmeland , Håkon , Luo X, et al. A Bayesian model for lithology/fluid class prediction using a Markov mesh prior fitted from a training image[J]. Geophysical Prospecting, 2019,67(3):609-623
Gong Q S, Huang G P, Meng X C, et al. Methods for lithology discrimination of volcanics in Santanghu Basin[J]. China Petroleum Exploration, 2012,17(3):37-41,6.
Xu D L, Li T, Huang B H, et al. Research on the identification of the lithology and fluid type of foreign Moilfield by using the crossplot method[J]. Progress in Geophysics, 2012,27(3):1123-1132 .
doi: 10.6038/j.issn.1004-2903.2012.03.037
Li W C, Yao G Q, Huang Y T, et al. Study on identification method of logging lithology for low resistivity reservoir in Wenchang 13-1 Oilfield[J]. Journal of Oil and Gas Technology, 2012,34(12):81-85,7.
Fan Y R, Huang L J, Dai S H. Application of crossplot technique to the determination of lithology composition and fracture identification of igneous rock[J]. Well Logging Technology, 1999(1):53-56,64.
Tian Y, Sun J M, Wang X, et al. Identifying reservoir lithology by step-by-step method and Fisher discriminant[J]. Progress in Geophysics, 2010,33(2):126-129,134.
Wang H, Li M B, Tang Y, et al. Tang YThe lithology prediction of ODP hole 1148A based on the wavelet neural network[J]. Progress in Geophysics, 2014,29(1):392-399.
doi: 10.6038/pg20140156
Wu S K, Cao J X. Lithology identification method based on continuous restricted Boltzmann machine and support vector machine[J]. Progress in Geophysics, 2016,31(2):821-828.
An P, Cao D P. Research and application of logging lithology identification based on deep learning[J]. Progress in Geophysics, 2018,33(3):1029-1034.
[14]
Corina A N, Hovda S. Automatic lithology prediction from well logging using kernel density estimation[J]. Journal of Petroleum Science and Engineering, 2018,170:664-674.
doi: 10.1016/j.petrol.2018.06.012
Yan J Y, Lyu Q T, Chen X B, et al. 3D lithologic mapping test based on 3D inversion of gravity and magnetic data: A case study in Lu-Zong ore concentration district, Anhui Province[J]. Acta Petrologica Sinica, 2014,30(4):1041-1053.
Liu Y X, He Z X, Zhang B T, et al. Integrated geophysical techniques for identification of igneous rocks[J]. Progress in Exploration Geophysics, 2006,29(2):115-118,5.
[18]
Paasche H, Eberle D. Automated compilation of pseudo-lithology maps from geophysical data sets: A comparison of Gustafson-Kessel and fuzzy c-means cluster algorithms[J]. Exploration Geophysics, 2011,42(4):275-285.
doi: 10.1071/EG11014
[19]
Deng C, Pan H, Luo M. Joint inversion of geochemical data and geophysical logs for lithology identification in CCSD main hole[J]. Pure and Applied Geophysics, 2017,174(12):4407-4420.
[20]
Konaté A A, Ma H L, Pan H P, et al. Lithology and mineralogy recognition from geochemical logging tool data using multivariate statistical analysis[J]. Applied Radiation and Isotopes, 2017,128:55-67.
doi: 10.1016/j.apradiso.2017.06.041
pmid: 28688247
[21]
Keykhay Hosseinpppr M, Kohsary A-H, Hossein Morshedy A, et al. A machine learning-based approach to exploration targeting of porphyry Cu-Au deposits in the Dehsalm district, Eastern Iran[J]. Ore Geology Reviews, 2020,116(C):223-234.
[22]
Rosenblatt M. Remarks on some nonparametric estimates of a density function[J]. Ann. Math. Statist., 1956,27(3):832-837.
[23]
E. Parzen. On estimation of a probability density function and mode[J]. Ann. Math. Statist., 1962,33(3):1065-1076.
[24]
Kazakos D. Choice of kernel function for density estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1980,2(3):255-258.
doi: 10.1109/tpami.1980.4767013
pmid: 21868899
[25]
Bolance C, Guillen M, Nielsen J P. Kernel density estimation ofactuarial loss functions[J]. Mathematics and Economics, 2003,32(1):19-36.
[26]
Fukunaga K, Hostetler L D. The estimation of the gradient of adensity function with applications in pattern recognition[J]. Trans Information Theory, 1975,21(2):32-40.