Automatic mud diapir detection using ANFIS Expert Systems algorithm; a case study in the Gorgan plain, Iran

Abstract Automatic seismic data interpretation is a significant method in oil and gas exploration. Seismic data interpretation makes the decision-making process difficult due to the complexities of the subsurface structures and the extent of related knowledge, and using the expert system method in seismic data interpretation can solve this problem. An expert system is a knowledge-based system that applies its knowledge in a complex and specific area and acts as an expert end-user consultant. This study includes investigating steps of designing an ANFIS expert system for mud diapirs detection with seismic data analysis, due to their importance in the exploration of hydrocarbon reservoirs in the high-potential region of Gorgan Plain. The expert system design process to identify mud diapirs on seismic sections was modeled in two approaches including manual and automatic seismic data interpretation. In the first approach, the experience of the expert was collected by manual interpretation of training data and used to create a knowledge base and inference of the expert system in the second approach. The validation verified the accuracy of this method with an average accuracy of 90.1% according to using minimum knowledge to develop a knowledge base of the designed ANFIS expert system.