Application of Genetic Artificial Neural Network in Diagenetic Reservoir Facies Study

Taking Fuyu oil-bearing layer of the north slope of Fuxin dome in Jilin Oilfield as an example,the pattern recognition based on genetic artificial neural network is used to study the diagenetic reservoir facies.Porosity,permeability,shale content and flow zone index are chosen to set up learning and predicting models of genetic artificial neural network.There are four diagenetic reservoir facies that are recognized: A-secondary pore diagenetic reservoir facies resulted from unstable components dissolution,B-mixture diagenetic reservoir facies resulted from medium compaction and weak-medium cementation,C-residual diagenetic reservoir facies resulted from strong compaction and medium cementation,D-micro-pore diagenetic reservoir facies resulted from strong compaction and cementation.A is the most promising diagenetic reservoir facies.