@article { author = {Mirzakhanian, Marzieh and Hashemi, Hosein}, title = {ANFIS Rules Driven Integrated Seismic and Petrophysical Facies Analysis}, journal = {Journal of the Earth and Space Physics}, volume = {47}, number = {4}, pages = {133-141}, year = {2022}, publisher = {Institute of Geophysics, University of Tehran}, issn = {2538-371X}, eissn = {2538-3906}, doi = {10.22059/jesphys.2022.331894.1007370}, abstract = {Different learning methods have been used to recognize seismic facies and reservoir characterization using seismic attributes. One of the significant issues in automatic facies analysis is to relate the seismic data to facies properties using the well data. According to previous studies, the role of attributes is more significant than the learning method for automatic classification. The proposed method uses supervised selection of seismic attributes for automatic facies analysis.Extended Elastic Impedances (EEI) at different angles as seismic attributes are being increasingly utilized in both seismic facies analysis and reservoir characterization. They are representative of elastic parameters of rocks appropriately. In the presented method, proper EEI seismic attributes are selected after a feasibility study using petro-physical logs, and EEI template analysis of the well data. Adaptive Neuro-Fuzzy Inference System (ANFIS) is applied to the fuzzy coded data of the well facies to train an automatic model to predict facies from the seismic data. Subsequently, the same particular EEI attributes are prepared. The EEI attributes from the seismic data are inputs for the trained ANIFIS model to perform seismic facies analysis. In this method, the seismic facies and the well facies are compatible. Only one well data can be sufficient for the well analysis stage and well facies clustering.The proposed method is applied on 3D prestack seismic data located in Abadan plain to discriminate hydrocarbon interval of Sarvak Formation. The results reveal that the supervised selection of attributes and fuzzy concepts present remarkable ability in dealing with imprecise seismic facies analysis and reservoir characterization.}, keywords = {seismic attributes,Extended elastic impedance,Facies analysis,Adaptive neuro- fuzzy inference system}, title_fa = {ANFIS Rules Driven Integrated Seismic and Petrophysical Facies Analysis}, abstract_fa = {Different learning methods have been used to recognize seismic facies and reservoir characterization using seismic attributes. One of the significant issues in automatic facies analysis is to relate the seismic data to facies properties using the well data. According to previous studies, the role of attributes is more significant than the learning method for automatic classification. The proposed method uses supervised selection of seismic attributes for automatic facies analysis.Extended Elastic Impedances (EEI) at different angles as seismic attributes are being increasingly utilized in both seismic facies analysis and reservoir characterization. They are representative of elastic parameters of rocks appropriately. In the presented method, proper EEI seismic attributes are selected after a feasibility study using petro-physical logs, and EEI template analysis of the well data. Adaptive Neuro-Fuzzy Inference System (ANFIS) is applied to the fuzzy coded data of the well facies to train an automatic model to predict facies from the seismic data. Subsequently, the same particular EEI attributes are prepared. The EEI attributes from the seismic data are inputs for the trained ANIFIS model to perform seismic facies analysis. In this method, the seismic facies and the well facies are compatible. Only one well data can be sufficient for the well analysis stage and well facies clustering.The proposed method is applied on 3D prestack seismic data located in Abadan plain to discriminate hydrocarbon interval of Sarvak Formation. The results reveal that the supervised selection of attributes and fuzzy concepts present remarkable ability in dealing with imprecise seismic facies analysis and reservoir characterization.}, keywords_fa = {seismic attributes,Extended elastic impedance,Facies analysis,Adaptive neuro- fuzzy inference system}, url = {https://jesphys.ut.ac.ir/article_85624.html}, eprint = {https://jesphys.ut.ac.ir/article_85624_fbeca5e3f8c9f9f5420eebac3e57fec2.pdf} }