Exploration, drilling, and production of hydrocarbons present unique measurement requirements. We will briefly describe each of these processes and the measurements used. We will also attempt to identify some of the larger themes that have influenced the direction of measurement technology for the oilfield. Exploration (i.e., finding oil and gas) involves assessing whether the right set of events have happened over geologic time to create a hydrocarbon-filled reservoir at a given location. This assessment involves a combination of geological studies, collection of geophysical surveys and the measurement of rock and hydrocarbon properties downhole and in laboratories. Most of the information we seek in exploration cannot be measured directly and must be inferred from related measurements. This reliance on inference has been handled in various ways, from geostatistics in the 1980’s to large-scale simulation and inversion in the 2000’s to the present explosion of machine learning applications. We look at how machine learning integrates measurements to answer exploration questions as well as to guide the design of downhole instruments to acquire only the most important data for inference.
Once a producible hydrocarbon reservoir is located, drilling a well requires two things: a rotating bit and circulating mud. The mud carries the “cuttings” or cut-away bits of rock out of the hole. The density of the mud is carefully controlled so that it exerts a pressure along the open well bore that is slightly greater than the pressures in the surrounding rock formations. If the mud becomes too light, downhole fluids will enter the well (called a “kick”). If the kick is not controlled it can lead to a blowout where downhole fluids flow uncontrolled out of the ground, often leading to explosions, injuries, and loss of life. A variety of sensors are used to help in detecting a kick, optimizing drilling penetration rate, navigating the bit along the desired trajectory, and characterizing the rock formations around the bit.
Once the well is drilled, production of oil (i.e., bringing it from the reservoir to the surface) needs to be carefully optimized to maximize the percent of oil that is recovered from the reservoir and minimize what is left behind. The field-wide optimization of production may be aided by measuring pressures and oil- water contacts in monitoring wells. Production logging tools, and more recently fiber optic sensors, monitor where flow is entering wells so that adjustments can be made to achieve a desired inflow profile. Volumes of gas, oil, and brine produced from each well or group of wells are often monitored at the surface. Producing oil too quickly from a well or from one point within a well will lead to early water breakthrough reducing the useful life of the well. There are maintenance issues as well. Corrosion and scale buildup inside wells and the performance of artificial lift systems that boost production from wells should be monitored. The cost and difficulty associated with obtaining downhole measurements is a major limitation for sensing and measurement to aid production. We look at some recent solutions to this high cost including permanently deployed sensors and small, untethered sensors.
Finally, we will reflect on the history of oilfield instrumentation development, identify some current trends, and speculate about future direction. At the conclusion of the tutorial, we will have some oilfield sensors and measurement tools on display and the scientists and engineers who designed them will be available to answer questions.
Max Deffenbaugh earned a B.S.E. in Electrical Engineering from Princeton University and M.S. and Sc.D. from the Massachusetts Institute of Technology / Woods Hole Oceanographic Institution Joint Program. He worked for ExxonMobil for 16 years at their Upstream Research Company in Houston, TX and their Corporate Strategic Research Laboratory in Annandale, NJ. In 2013, he joined Aramco Services Company to create and lead their Sensors Development Team in Houston, TX. His interests include research, development, and commercialization of sensors and measurement tools for the oilfield.
Chinthaka P. Gooneratne received the B.Eng. (Hons) degree in Information and Telecommunication engineering, and the M.Eng. degree in the area of Electromagnetics, from Massey University, New Zealand, in 2004 and 2005 respectively. He was awarded a Monbukagakusho scholarship from the Government of Japan in 2006 and earned a PhD in Electrical Engineering in 2009 from Kanazawa University, Japan. He currently works in the Drilling Technology Division of EXPEC ARC, Saudi Aramco on sensors and instrumentation for drilling & workover applications.