Predictive Systems for the Well Drilling Operations

Abstract

AbstractThe article outlines the potential for the development of predictive systems for the drilling operations, as well as the main criteria that will determine technical requirements for research and development in these areas. Operational drilling efficiency improvements would include a whole range of options from developing solutions to decrease non-productive time (NPT) (e.g. by selecting optimum drilling parameters) to equipment performance analysis in order to predict equipment failures. The key areas where digital prediction systems could be implemented are the following: remote monitoring of drilling operations; predictive rig maintenance and repair system; predictive NPT recording and analysis system; geosteering, geomechanics, and petrophysics soft-ware; drilling failures prediction and prevention during drilling oil and gas wells; automated drilling system; and well-targeting technology system. The standalone software packages are to be combined into one automated and adaptive decision-making system based on physical and mathematical models of well construction and rock conditions. The presented study summarized a minimum number of models necessary for the operation of a predictive drilling system. The practical implementation of digital solutions in drilling is analyzed through the main approaches used in the development of predictive analytics algorithms. The basic assumptions for all algorithms are formulated to ensure the limitations of the developed model. Also, technical requirements are formulated for a uniform data collection and storage system that is the foundation of the predictive analytics block.KeywordsOil and gas well drillingPrediction systemDrilling decision making systemBig dataArtificial intelligenceIndustrial IoT (internet of things)