Technical Report: Microsensor Development and the Mapping of Wormholes by the PTRC’s EOR Research Network
Executive Editor’s NoteWe continue the new section we started in our March 2013 issue of JCPT informing our readers of the new and exciting research that is being conducted in various Canadian research organizations. As you know, all the technical papers published in the journal are peer reviewed. The primary principle behind the SPE peer-review process is to provide a fair and timely technical review to identify the most meaningful technical papers that contribute to the development of petroleum technology.
However, not all significant pieces of technical work are submitted for peer review or have all the results in yet. There are many good-quality relevant technical reports produced by non-profit Canadian research organizations that would benefit our readers. We continue to highlight some of the research conducted by these organizations with Petroleum Technology Research Center (PTRC) this month. I should emphasize again that these highlights are not peer reviewed by JCPT, but we think that they are of interest to our readership.
I hope you enjoy reading about the research on mapping of wormholes with miscrosensors, and if you are interested, you can investigate further by contacting PTRC. We also welcome your feedback regarding the new initiative as well as any other aspect of JCPT.
BackgroundThe Petroleum Technology Research Centre (PTRC), based in Regina, Saskatchewan, was founded in 1998 by the Governments of Canada and Saskatchewan, the University of Regina, and the Saskatchewan Research Council to fund research into enhanced oil recovery (EOR) with a particular focus on heavy-oil deposits (which proliferate along the Alberta-Saskatchewan border). The organization is a not-for-profit corporation that fosters research and development into EOR and carbon storage, with the goals of improving recovery rates while reducing the environmental footprint of the oil and gas industry. The PTRC has managed an EOR research program since its inception, focussing in particular on cold heavy-oil production with sand (CHOPS), which makes up the majority of heavy-oil production in western Canada. This process involves the production of large amounts of sand in suspension with the oil, which is later separated out. This production of large amounts of sand leaves voids in the reservoir. These voids appear to cause increased connectivity between wells and decreasing reservoir pressure, rendering waterfloods ineffective. These voids have been termed “wormholes” by industry. If more was known about the nature of wormholes, their morphology, size, direction, and path, better decisions could be made to maximize sand and oil production.
In 2011, the PTRC, along with its Dutch partner, INCAS3, set up a joint project with its industry members to develop a technology that would enable small autonomous sensors to flow through a heavy-oil reservoir (Smith 2013). The proposed wireless sensor network would act as spatially distributed autonomous sensing “motes” to monitor environmental conditions or parameters (e.g., temperature, vibration, pressure, motion, chemicals, gases, or pollutants). Motes would probe these candidate parameters and then cooperatively pass probed data through the network to a main location. In some cases, probed data would be harvested after recouping motes from their environment. Essentially, each sensor or “mote” is an autonomous pod that communicates with the mote adjacent to it, and with a swarm of hundreds or even thousands of motes, they could communicate in a chain back to a central receiver. Another option is to have a retrievable mote that records its positioning and other data onboard, to be downloaded later after retrieval.
It was an ambitious project, since off-the-shelf sensors are unable to currently perform these tasks at the small sizes required to flow through a reservoir. It was quickly realized that a staged approach to the problem would be necessary. With the PTRC providing oilfield expertise and access, and INCAS3 providing the sensor development knowledge, the project could begin with trials of empty or “blank” motes. Essentially, these are empty plastic hulls of various sizes and shapes to determine the largest possible sensor that could travel intact through a reservoir, and associated pumps and piping.
The Pump TestThe progressive cavity pump is what makes CHOPS production possible. The mechanism within one of these pumps is a large screw “rotor” spinning in a spiraled “stator.” This posed potential problems in the recovery of sensors. While screw pumps such as these are effective in bringing up sand and oil, they may destroy 5 to 7 mm plastic hulls. In July of 2012, the first test of blank motes was performed. A horizontally mounted PCP pump with the same specifications as the pump in the future field site was chosen. A total of 45 motes of varying size (5 to 12 mm) and shape (spherical, elongate, and cubic) were sent through the pump at a pressure of 55 to 69 kpa. During this test, it was discovered that motes of up to 9 mm in size can flow through a progressive cavity pump without being ground up, although some of them did break along a seam upon collision with the top of the pump housing. This highlighted the need for strength in hull design in the unforgiving environment of oilfield reservoirs and production equipment. Special testing was undertaken in the Netherlands to verify the pressure that each mote would be able to withstand. Each mote was designed and built to withstand 1,450 psi (10 MPa) without being crushed. Although in this test, some of the motes did develop a leak but the instruments inside were undamaged. This is a significantly higher pressure than would be expected in a heavy-oil reservoir in western Canada.
The Field TestFollowing the pump and pressure tests, it was time to move the entire project to a field-test site. A site owned by one of the members of the PTRC enhanced-oil-recovery consortium was chosen for the initial field trial of the blank motes. The site in northern Alberta included a pair of wells—a producer and an injection well both drilled directionally away from each other at an angle of 47° to a distance of 345 m apart in the reservoir. The top of the producing zone was at an average depth of 325 m true vertical depth. This was previously the site of a waterflood operation that had been shut in because of lack of production and breakthrough between the two wells.
The project objective was to gauge the ability of the different-sized sensor motes to travel through the CHOPS reservoir – both in terms of the number of motes successfully going from injection to production well, and their condition upon completion of the path through the reservoir. The first activity involved putting dye into the injector at the same time as the first set of motes. The operation of both the injector and producer would continue as normal until evidence of motes or dye came through a filter screen at the producer. This operation was fraught with many challenges, aside from the obvious one of solid particles flowing through a reservoir. Neutral density of the motes was important—too heavy and they fall into the sump of the injector (or producer), too light and they never get flushed into the perfs. For this test, some of the motes had and RFID tag placed within them, and are referred to as being “instrumented” so they would be sensed by receivers on the surface to log the time of injection and production exactly. Some of them were hollow hulls, and some were solid plastic spheres and capsules.
A total of 25,258 blank motes were injected for this test. 23,978 of them were solid green and pink spheres of 5mm and 7mm diameter with a specific gravity of 1.04. 400 more hollow 5mm capsule and sphere shape motes were also injected. These had specific gravities of 1.04 and 0.94 in an effort to determine which density was optimal for recovery and transit in the wormhole network. 880 “instrumented” (RFID tagged) motes of 5, 7, and 9 mm capsules and spheres were also injected. For the initial test, motes were injected simultaneously with dye. Dye was produced after approximately 90 minutes, but no sign of motes was observed. At the end of the first day, the injector was turned off, and the producing well continued to pump.
The next morning, 88 motes were found in the screen (Fig. 1), some of them instrumented and many of them not, and the ratio of green-to-pink-coloured motes was identical to those that were injected.
The test carried on for 3 more days, varying the injection and production rates, and the volume and types of motes. In the end, with more than 25,000 motes having been injected, 1,149 were retrieved in various states of deformation from only slight, to more than 50% deformed. Interestingly, many of the 7-mm motes were recovered, allowing for more internal space when full instrumentation of the sensors becomes possible. The motes spent much more time in the reservoir and wells than did the dye with one mote spending more than 38 hours in transit; this was likely caused by the irregular morphology and pathways of the wormholes in the reservoir. This is supported by the results showing that none of the larger mote-sizes (9 mm) were produced but remained in the reservoir. Much later, after bailing out the injector well, many more motes were found that never entered the reservoir in the first place. The end of the first field test led to the following conclusions:
- To the best of our knowledge, it is the first experimental proof that a solid object of several mm-size can pass through a heavy-oil reservoir.
- Motes needed significantly greater time to pass from the injection well to the production well than dye.
- In total, 7.6% of injected sensor motes were retrieved from the reservoir.
- Small sensor systems can be sent into, and retrieved from, a wormholed reservoir after travelling hundreds of metres.
- There is at least one channel in the wormhole network with a diameter of at least 7 mm, which connects the injector and the producer directly.
- The well equipment was not damaged by sensor motes.
- Producing motes with exact specific gravity is not critical in terms of the motes being too heavy or too light.
- For this particular site, optimal motes can be 7-mm spheres and 7-mm capsules.
Next StepsThe PTRC, in conjunction with INCAS3 and industry partners, is continuing to move forward with the development of smart sensor systems under the explicit goal of the characterization of wormhole networks in CHOPS producing areas. Field tests of blank motes are planned for different reservoir areas and injector/producer configurations. For the first time, camera-equipped sensors will be sent into a reservoir and live pictures of a wormhole may be obtained. The work on the development of communicating sensor arrays of the size and shape suitable for reservoir mapping continues in earnest at some of the most cutting-edge laboratories in the Netherlands.
To become a member of the PTRC heavy-oil consortium, (HORNET) and have access to the results and planning of not only this exciting work, but many other lines of research, please contact email@example.com and visit our website at http://ptrc.ca.