Advancement and economic benefit of geosteering and well-placement technology

Abstract

In the development of many of today's reservoirs, the oil and gas industry is challenged to drill more efficiently and is asked constantly to maximize recovery and production. However, drilling through these reservoirs is challenging because geologic models often are limited to the resolution of seismic data, and offset wells often have significant variations. Geosteering is the process of adjusting the borehole trajectory in real time to correct for unanticipated variations in geology and structure to avoid exiting the target zone. It is a technique currently used on many horizontal and deviated wells for better well placement and for efficiently draining a reservoir. Recent improvements in well-placement and formation-evaluation technologies have helped in gaining access to bypassed reserves that originally were not thought to be practical targets. Examples highlight the economic benefit of geosteering and well-placement technology. Maximum reservoir contact in the sweet spot leads to increased production, and early warning of approaching faults and bed boundaries results in reduction of sidetracks. Furthermore, keeping the well trajectory away from the oil-water contact optimizes production by producing less water. Finally, maximizing production by placing the wellbore entirely within the best reservoir zone boosts productivity so that wells that previously appeared difficult or uneconomic are now becoming viable.

A Novel Method for Locally Updating an Earth Model While Geosteering

“…The recently developed ultra-deep directional electro-magnetic (deep EM) technology is sensitive to resistivity contrasts up to tens of meters around the wellbore (see e.g. [1] [2] [3]). Deep EM measurements is a technological step change that gives more information about geological structures, fluid contacts and reservoir properties deeper into the formation than traditional Logging While Drilling (LWD) measurements, even ahead of the bit [4].…”

“…Two properties, represented by the two property functions Φ 1 and Φ 2 for the two layers L 1 and L 2 respectively, are linked with each region using four corresponding mappings { } f . Informally, we can say that the mapping deforms the interior of 1 1 D into the shape of 1 1 R . This allows the subgrid to be populated by sampling values from Φ 1 .…”

“…Informally, we can say that the mapping deforms the interior of 1 1 D into the shape of 1 1 R . This allows the subgrid to be populated by sampling values from Φ 1 . This deformation is further explained in Section 5.…”

“…The effect of deforming e.g. 1 1 D into the shape of 1 1 R is that it is elongated in horizontal direction and squeezed together in vertical direction. The white near-vertical In the simple example in Figure 2, created using a rudimentary software prototype, the property functions used to populate each layer are identical.…”

“…For linking the property function Φ for a layer with the region R, we apply a mapping f. First, let us assume that we have an unfaulted layer L, represented by a single region R. The exact shape of L is a matter of geological interpretation, so we assume that we have arrived at an alternative we call L 1 . An informal and intuitive way to understand the mapping of properties from a property function to a layer is to imagine that the shape of Φ is deformed into the shape of L 1 as shown in Figure 3.…”