Investigation of the role of quantum sensors and robotic platforms in improving the accuracy of forecasting hydrocarbon reserves in offshore zones

Abstract

The present study is devoted to the development and testing of an integrated methodology for the use of quantum sensors and autonomous robotic platforms to improve the accuracy of geological forecasting of hydrocarbon reserves on the continental shelf. A comprehensive assessment of the effectiveness of new-generation quantum gravimetric and magnetometric systems with sensitivity up to 10-12 g/√Hz, integrated with autonomous underwater vehicles equipped with artificial intelligence systems for adaptive data collection, has been carried out. Experimental studies were performed on the basis of the Prirazlomnoye and Shtokman deposits in the Barents Sea using multispectral analysis of geophysical data. The results demonstrate an increase in the accuracy of volumetric modeling of hydrocarbon structures by 37.6% compared with traditional methods of seismic exploration. The developed neural network forecasting model, which integrates quantum sensor data and a multiparametric analysis of geological structures, reduces uncertainty in estimating reserves to 8.3%, which significantly exceeds existing methodologies (standard error of 19.4-22.7%). The technical and economic analysis indicates a 42.3% reduction in exploration drilling costs when implementing the proposed approach. The presented methodology opens up prospects for the development of digital technologies in marine geological exploration and provides a qualitatively new level of predictive accuracy in the study of complex hydrocarbon formations of the continental shelf.