Research On Sedimentary Facies Of Semantic Classification And Knowledge Database Based On Ontology

Sedimentary facies is the basis of oil and gas storage research,and has guiding significance for predicting sand body distribution and reservoir physical properties.Single well phase analysis is one of the key aspects of sedimentary facies research,and the identification of sedimentary microfacies using typical well logging calibrated by core is the most common method for single well phase analysis.However,there are two problems in relying on manual identification of the logging phase and interpretation of the sedimentary microfacies.First,the workload is large,the experience of the researchers is high,and the results are subjective.Second,the logging data is incomplete and there is a missing The phenomenon of item feature data.Therefore,problems in the field of sedimentary facies can be solved by means of ontology and semantic classification techniques.Ontology,as a conceptual model modeling tool,can describe information systems at the semantic and knowledge level and has been widely used in the field of knowledge engineering.Therefore,it is feasible and beneficial to introduce bulk technology into the field of sedimentary facies.The work of this paper begins with the construction of sedimentary facies ontology model,and on this basis,the two parts of sedimentary facies semantic classification and knowledge base research are studied.In terms of semantic classification,based on the constructed sedimentary facies ontology,combined with information volume and semantic distance,this paper proposes a semantic classification method for braided river delta sedimentary facies based on ontology information.Among them,the core of the method is to use the amount of ontology information to measure and represent the relevant features of the sedimentary facies.Then,through the semantic distance formula of the design,the semantic similarity is obtained,and the sedimentary facies identification classification is realized.The method optimization is mainly based on the characteristics of Minkowski distance and Euclidean distance,the semantic distance formula is optimized,the matrix form conversion formula is used to obtain the semantic coefficient matrix,and the coefficient adjustment formula based on the classification result feedback is set,thereby achieving the method tuning.The experimental verification of the actual oilfield data shows that the method can achieve the high classification accuracy of 85.42%,instead of manually implementing the sedimentary facies classification,which provides a highly efficient alternative method for the actual sedimentary facies division.In the knowledge base,a knowledge base framework based on sedimentary facies resource types was designed by summarizing the characteristics of sedimentary facies.Based on this,a sedimentary facies knowledge base was constructed.The designed knowledge base framework follows three principles of full coverage,reusability,and generalization.Therefore,this paper divides the sedimentary phase knowledge base into three parts: knowledge layer,data layer and picture layer.Among them,the knowledge layer is the core part of the entire knowledge base,which is based on the constructed sedimentary facsimile ontology and constructed using Jena and other technologies.The designed sedimentary facies framework provides the possibility to realize the knowledge base.The preliminary constructed knowledge base provides a knowledge sharing platform for relevant geological researchers and lays a partial foundation for further large-scale geological knowledge base construction.