Title :
Identification of metamorphic rocks in the CCSD main hole
Author :
Pan, Heping ; Luo, Miao ; Zhao, Yonggang
Author_Institution :
Inst. of Geophys. & Geomatics, China Univ. of Geosci., Wuhan, China
Abstract :
It is more difficult to identify metamorphic rocks than sedimentary rocks by well logs. In order to identify metamorphic rocks, eight types of well logs were chosen for the identification of metamorphic rocks in China Continent Science Drilling (CCSD) main hole. We use stepwise discrimination method to recognize metamorphic rocks. The first step is to establish Back-Propagation artificial neural network model to recognize general types of metamorphic rocks. Then the Bayes discrimination function was established to recognize detailed types of metamorphic rocks.
Keywords :
Bayes methods; geophysical techniques; neural nets; rocks; Bayes discrimination function; CCSD main hole; China Continent Science Drilling main hole; back-propagation artificial neural network model; metamorphic rocks; sedimentary rocks; well logs; Artificial neural networks; Drilling; Geology; Geophysics; Neurons; Neutrons; Training; Bayes discrimination; CCSD; Identification; artificial neural network; metamorphic rocks;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Conference_Location :
Yantai, Shandong
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5584844