Search

In this paper, a new series of Logging-while-drilling (LWD) nuclear magnetic resonance (NMR) tools is introduced. The tools incorporate a magnet arrangement which provides a low field gradient minimizing adverse lateral motion effects. T2 distributions are measured while drilling or sliding in a range of hole sizes, including large holes previously inaccessible to LWD NMR measurements. Benefits and trade-offs of key tool design features are presented. Emphasis is placed on measurement quality, operational simplicity and log quality control in real time and memory mode through example logs from operator wells. Real-time T2 distributions provided by the new generation LWD NMR tool enable a full range of NMR answers, including lithology-independent total porosity, bound fluid volume and permeability. Real-time log quality controls monitor tool noise, antenna sensitivity, tuning quality and lateral motion. Acquisition sequences are optimized for porosity precision and accuracy in different environments. Drilling data have been acquired in several operator wells and a test well covering diverse ranges of formations and logging conditions. Environmental factors affecting porosity precision and T2 quality are discussed with reference to log quality indicators and comparison with wireline NMR logs is made where available. In general, the LWD tools deliver NMR answers comparable to the analogous wireline logs in terms of precision, accuracy and vertical resolution. With increased industry focus on LWD services and heightened sensitivity to downhole chemical sources, the need for reliable LWD NMR measurements continues to grow. In addition to a sourceless porosity measurement, NMR provides unique quantitative information on the disposition of producible fluids, which is not available from other logs. Real-time producibility information can be used to optimize formation pressure measurements and sampling as well as for timely completions and well placement decisions. The new generation LWD NMR tools introduced here addresses this need.

This account describes how advanced well placement technology helped to optimize horizontal well position and maximize hydrocarbon production in deep water turbidite reservoirs. The deep reading directional electromagnetic tool, a latest-generation LWD (Logging While Drilling) measurement, was the technology differentiator for optimizing well placement in a number of deep water horizontal wells. The new directional measurement is highly sensitive to reservoir boundaries and therefore gives early warning of conditions requiring steering adjustments while drilling horizontal wells, maximizing well position in the reservoir. This paper shows how thin oil rims, faulted reservoirs and those with highly variable structure were able to be developed more efficiently. By reducing uncertainties about the reservoir, the new technology helped optimize production, eliminate sidetracks and minimize well construction cost and risk. One of the main challenges of maintaining a horizontal wellbore inside a thin hydrocarbon bed is the uncertainty of formation dip. Inside a hydrocarbon bed, real time resistivity images often yield dips that are more representative of stratigraphic features rather than structure defining reservoir boundaries (from surrounding, conductive shale beds). The deep-reading azimuthal electromagnetic measurement helped in early detection of neighboring conductive beds, and the distance to these beds. This ensured that the wellbore was drilled parallel to the structure and inside the ‘sweet spot’ away from non-productive layers. As reservoirs become more challenging to develop, the use of advanced technology often helps minimize risk. When fields are developed with horizontal wells, it is desirable to maximize well position inside the reservoir layer since the rate of increase in hydrocarbon production increases as horizontal length in the reservoir increases. We show an example of how advanced well placement technology was successful in doubling hydrocarbon production rate compared with wells in the same field drilled with conventional technology.

A UAE offshore operator drilled a pilot well in a complex layered Middle Jurassic formation to confirm where to land a high inclination production drain.(fig. 1) The reservoir underwent gas injection EOR. Reservoir simulation (fig. 2 & fig. 3) predicted possible gas fingering that required further confirmation before committing to drill the production drain. The objective was to positively identify all the intervals where liquid hydrocarbons could be produced avoiding the dry gas zones. The petrophysical analysis was complicated by rock heterogeneities and fluid mixture including gas, heavy and light oils plus water. A traditional approach based on standard logs was not conclusive. The addition of NMR data versus depth allowed carbonate texture differentiation and a derivation of a simultaneous multi-fluid solution that highlighted the zones with possible oil production. This initial information was used to select suitable intervals where to perform nuclear magnetic resonance stations with diffusion editing sequences designed for maximum sensitivity in the gas to light-hydrocarbon region. Several stations were acquired with success and processed in pseudo real time making use of a VSAT link to share data with wellsite and predict the most likely downhole fluid composition. With integrated depth and stationary information available, a few sets of sampling stations with real time downhole fluid identification were chosen to verify dynamically which formation fluid would finally flow. Sampling was achieved using either a single probe or dual packer, depending on rock permeability. The fluid was analyzed in-situ using an innovative downhole fluid separation technique and some samples were taken to surface for a final PVT confirmation. The combined information proved to be very consistent and provided a detailed reservoir description inclusive of fluid types.(fig. 31, fig. 32 & fig. 36) The immediate impact was the change of the target drain location to a deeper zone in the reservoir structure. In fact, the zone a priori selected would have produced dry gas, due to the poor miscibility of injected gas and relatively viscous hydrocarbons in place. The methodology used was well communicated to the various operator departments and had a broad acceptance, opening the way for a systematic similar approach in the future.