Decision Making for Reduction of Nonproductive Time through an Integrated Lost Circulation Prediction

An increase in world energy demand forces oil and gas companies to drill deeper in order to produce more oil and gas. This requires drilling layers with various characteristics and results in drilling problems as drilling progresses. Lost circulation and stuck pipes are the most common drilling problems, which are always challenging. Mechanical pipe sticking is likely to occur after complete loss. Time that is spent dealing with such problems is called nonproductive time. It is also defined as time during which drilling is ceased or the penetration rate is very low. This definition includes consequences of lost circulation. Lost circulation is likely to take place throughout drilling operations and even during primary cementing due to direct contact of the drilling fluid with the formation along with severe pressure pulses due to pipe movement or onset of circulation after connection, which sometimes goes over hundreds of psi. Lost circulation results in a loss of money, especially when severe losses are experienced. Partial mud loss, or even complete mud loss, can occur in low fluxes, which can lead to well blowout or severe pipe sticking. Freeing the pipes may involve a week or even more time from the rig. Thus, accurate information on returned fluid and recording mud loss rate can help prevent drilling problems. A recent solution to dealing with lost circulation is an underbalanced drilling technique, although this is not applicable in some countries due to lack of technology or great expenses. Thus, prediction of loss severity can bring the opportunity of decision making true for adjusting drilling fluid content and operational parameters. Several factors while drilling will govern how severe mud loss would occur. These make analytical modeling of lost circulation complicated. Employing artificial intelligence can be a leeway with proven capability and accuracy. In this article, the operational parameters in a Middle Eastern oilfield are used for prediction of the mud loss severity along different sectors of the oilfield. Cross-validation showed fairly good compatibility with reality.