DocumentCode :
2519533
Title :
A new method to predict the reservoir porosity based on fuzzy-PCA and neural network
Author :
Ma, Zhenglie ; Luo, Xiaoping ; Du, Pengying ; Hou, Jiagen ; Duan, Dongping
Author_Institution :
Coll. of Electr. Eng., Zhejiang Univ., Hangzhou, China
fYear :
2011
fDate :
23-25 May 2011
Firstpage :
2587
Lastpage :
2590
Abstract :
Porosity and permeability are the two fundamental and crucial reservoir parameters which are often used in reservoir description. However, the two properties are difficult to be measured and predicted, due to some influences such as rock type and cement and so on. In this paper, we proposed a new method combined of fuzzy theory, principal component analysis and the neural network to predict the porosity by well log in Yangerzhuang oil field. The experiment results demonstrate that the method in this paper can retain the information more effectively in the process of dimension reduction, and thus greatly the prediction accuracy can be improved.
Keywords :
fuzzy set theory; hydrocarbon reservoirs; neural nets; permeability; porosity; principal component analysis; well logging; Yangerzhuang oil field; fuzzy theory; neural network; porosity; prediction accuracy; principal component analysis; reservoir permeability; reservoir porosity; well log; Artificial neural networks; Covariance matrix; Eigenvalues and eigenfunctions; Neurons; Noise; Principal component analysis; Reservoirs; Dimension Reduction; Fuzzy theory; Neural Network; Porosity; Principal Component Analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control and Decision Conference (CCDC), 2011 Chinese
Conference_Location :
Mianyang
Print_ISBN :
978-1-4244-8737-0
Type :
conf
DOI :
10.1109/CCDC.2011.5968647
Filename :
5968647
Link To Document :
بازگشت