Transition to the use of digital assistants in the kinematic interpretation of the data of seismic exploration by the example of the problem of improving the quality of seismic data after summation and reliability of the tectonic model forecast

Modern seismic exploration still faces the challenges of automating processes and increasing the reliability of work results, especially in regions with complex geological conditions. An important place in the cycle of seismic surveys is occupied by the stage of kinematic interpretation, the main purpose of which is a detailed understanding of the structural features of the geological section and obtaining a reasonable geological model of a particular region of study. The cost of an error at this stage of the work is quite high, but the interpretation processes require significant labor costs, and the results often contain errors. Standard algorithms and methodological approaches do not fully provide solutions to the full range of tasks, which necessitates the search for new approaches to the interpretation of seismic data. In recent years, there has been increasing interest in attracting the capabilities of artificial intelligence to solve production problems. New approaches to solving the problems of the stage of kinematic interpretation of seismic data based on the use of artificial intelligence through machine learning and deep neural networks are proposed: – technology of elimination of irregular noises of the total seismic data to improve the quality of the initial seismic material and simplify the stage of structural interpretation; – technology of probabilistic forecast of disturbance systems and obtaining a detailed tectonic model. Theoretical foundations are presented and the results of applying technologies on a series of real production projects are demonstrated, which confirm the advantages of using neural networks in interpretation to eliminate subjectivity and significantly reduce time costs at the stage of structural constructions in various geological conditions.

seismic interpretation, tectonic faults, noise effects of seismic recording, tectonic model, automation, artificial intelligence, machine learning, deep neural networks, methodology