Table of Contents

Vol. 7, No. 2
2nd Quarter 2001

Field Results from DOE's Independents Projects

DOE's National Petroleum Technology Office (NPTO) made multiple awards during 2000 in its "Technology Development with Independents" projects. Results from one project were published in PTTC/World Oil's Petroleum Technology Digest section (March 2001 issue, accessible at http://www.pttc.org) and results of two more are summarized herein. Results from even earlier projects are accessible at http://www.npto.doe.gov/indep/TOC.html

Awards from Phase 1 of a 2001 solicitation are anticipated soon, and the due date for Phase 2 and 3 proposals is August 15 and December 24. Maximum award for these short-term projects will be $75K and a minimum cost share of 50% is required.

Pilot J-Sand Waterflood, Nebraska. Coral Production Corp. (Coral) operates wells producing by primary drive from the 5,300 foot J-Sand in the Herboldsheimer Field (now unitized into the Cliff Farms Unit) in the Denver-Julesberg Basin in the Western Nebraska Panhandle. Since waterflooding failures are common due to channeling, high gas saturation, wettability or fractures, Coral needed extra confidence before implementing a waterflood. As is typical, Coral did not have access to the detailed data required to develop a reservoir model and simulate performance. However, useful field information was available in the public domain. 

Coral retained the Correlations Company, a software/technology consulting firm in Socorro, NM, to apply fuzzy logic and neural network techniques. They compiled and reviewed data from 140 J-Sand waterflood projects. Using the Secondary to Primary (S/P) ratio as the independent variable, they looked for correlations with the dependent parameters (parameters logically known to influence waterflood success). As expected, initial analysis using conventional cross-plots failed, but efforts were successful using fuzzy ranking and neural network techniques. A S/P ratio of 3.16 is estimated for the field, which equates to more than 100,000 bbls of expected secondary recovery. 

Encouraged by this result, Coral evaluated available pressure data, identified a prospective area, and initiated a pilot waterflood in mid-2000. Data to confirm reliability of the predicted waterflood success are still being collected.

Increased Reserve Potential in Queen Sand Waterflood, SE New Mexico.  MNA Enterprises LTD, Co. (MNA) recently acquired the 520-acre Shugart Queen Sand Penasco Unit. The field was developed in 1939 and had been under waterflood since 1974, although water injection operations had been suspended. Fourteen wellbores have been drilled through the 3,350-foot Queen Sand with the 11 remaining wells producing only 100 bopm (plus 300 bwpm). With a Secondary to Primary (S/P) ratio of 0.7, as compared to the average ratio of 1.2 experienced in area waterfloods, MNA suspected that the prior waterflood had been prematurely terminated. 

MNA performed a detailed analysis, consisting of a data acquisition and reservoir characterization phase, production history match using DOE's BOAST III reservoir simulation model, and performance predictions for different operating scenarios. To compensate for extremely limited log data within the Unit, the study area was expanded to encompass four contiguous sections. Modern logs from non-Unit wells drilled to deeper horizons were used to correlate older Unit logs using neural network approaches. Mapping software was used to interpolate and map reservoir properties required as simulation input. 

Modeling indicated that simply re-initiating water injection at 300 bwpd would economically recover an additional 7,000 bbl of oil. But, if the pattern were realigned by converting one producer to injection and one injector to production, sweep efficiency would improve, recovering an additional 36,000 bbl of oil. Further, adding another producer would yield yet another 15,000 bbl of oil. The model results also indicated that if additional injection capacity could be developed, to 600 bwipd, oil production rates could be doubled. The characterization effort identified potential in a deeper, previously undeveloped Penrose zone and a possible combination Queen sand/Penrose drilling location was identified.

Neuro3. Neural network technology is a common element to both projects. DOE's NPTO has developed "Neuro3," a neural network software package as part of its Risk Analysis toolset. Neuro3 uses the backpropagation neural network (BPN), the most widely used feedforward neural network system. Backpropagation refers to the training method by which the weights of the network connection are adjusted. The calculation procedure is feedforward, from input layer through hidden layers to output layer. During training, the calculated outputs are compared with the desired values, and then the errors are backpropagated to correct all weight factors. Neuro3 is a 32-bit MS Windows application. It contains a help system with a tutorial and background information, plus it has a spreadsheet interface to allow import and export of external data sets. 

For information about the projects with independents, contact DOE's Jim Barnes, phone 918-699-2076, email Jim.Barnes@npto.doe.gov or Walt North, phone 918-699-2026, email Walt.North@npto.doe.gov. For more information about Neuro3, contact Chandra Nautiyal, DOE's Project Manager, phone 918-699-2021 or email Chandra.Nautiyal@npto.doe.gov.

Stripper Well Consortium Funds 13 Projects

Soon after their April 9&10 meeting in Pennsylvania, where they reviewed 23 R&D proposals, the Executive Council of DOE's Stripper Well Consortium (SWC) approved funding for 13 projects, committing $912,000 of DOE funding. Topics for these short-term, field-oriented projects include: waterflood improvement, identifying candidates (either bypassed or damaged) for stimulation/restimulation, artificial lift, and environmental-related. A majority of the projects are in the Appalachian area, but there are projects in the Rockies, Oklahoma, New Mexico and Texas, and findings from most of the projects should be broadly applicable. Work within these projects is expected to start quickly. 

For further information about the individual projects, visit SWC's website (http://www.energy.psu.edu/swc/).

Top of Page

DOE Stripper Gas Well Projects Explore Different Approaches for Identifying Under-Performing Wells

Three of four stripper gas well projects funded by DOE explore different approaches for identifying under-performing wells. Although performed in a certain area, conclusions have validity for stripper well operators in other locations. Brief descriptions of progress in the three projects follow.

Using Three-Month Data (Ohio). James Engineering, Inc. (James), working with Artex Oil Company, evaluated gas production declines and basic well histories of 376 wells located primarily in Ohio that produced from the Clinton formation. Within this group, 72% of the wells experienced abnormal production decline, defined as a consecutive three-month period where actual production fell below a type-curve forecast by 50% or more. The wells with abnormal production decline were then categorized into one of eight categories for statistical analysis.

Statistical analysis showed that three problems - fluid accumulation (46%), gathering system restrictions (24%), and mechanical failures (23%) - accounted for more than 90% of abnormal production declines. Prior perceptions were that many abnormal production declines were related to reservoir damage or precipitate plugging. Solving these simpler problems, which occurred in the vast majority of the wells, is considerably lower cost. James has developed a preliminary triage and well analysis form to assist operators in gathering the information to detect and correct abnormal production declines. Project work will be finished in the summer of 2001.

SWARM Software-Using Cumulative Production Data (Pennsylvania). Schlumberger Holditch-Reservoir Technologies developed a new PC-based software product, SWARM - Stripper Well Analysis for Remediation Methodology. The software quickly compares individual well performance, using cumulative production data, with that of adjacent wells. The comparison takes into account depletion due to the variable date of first production. The operator pre-selects the cumulative production period and the radial distance to surrounding wells. Required input data are primarily the well's location (x:y) and production history. Final work is underway to complete the user's manual and the data import interface. Work should be completed within 12 months. 

The software was beta-tested in a stripper gas well field in northeastern Pennsylvania. Quickly screening over 700 wells, SWARM identified wells that stood out against adjacent wells. For example, one well drilled in 1988 had cumulative production after eight years of 51 MMCF, while five other wells within 4,000 feet drilled in the 1986-1992 time frame had eight-year cumulatives ranging from 148 to nearly 300 MMCF. This well obviously warrants further attention.

Using Type Curve Analysis (Oklahoma). Advanced Resources International (ARI), working with Oneok Resources, is using type-curve methodology to analyze approximately 100 wells in the Mocane-Laverne field in Oklahoma. Production is from the Morrow, Chester, Hoover and Tonkawa. Since the Morrow is the second-largest gas play in Oklahoma and Mocane-Laverne is the largest Morrow field, results should be broadly applicable across the Midcontinent. ARI is soliciting research partners for an additional site.

In previous studies, ARI has developed a methodology for identifying restimulation candidates using a combination of production statistics, production type-curves, and virtual intelligence techniques. In this effort, they are extending the methodology to not only identify restimulation candidates, but to also identify other types of production enhancement opportunities.

The methodology will utilize production type-curves to predict reservoir properties and completion effectiveness (i.e. permeability, fracture half-length, and/or skin damage). Then, artificial neural networks will be used to build well performance models and identify well and reservoir performance drivers (i.e., which stimulation or completion techniques provide the best well performance). In addition to neural network modeling, a variety of virtual intelligence methods will be performed, in combination with type-curve analysis, to select candidates for remediation.

A "Stripper Gas Well Enhancement" manual is being developed. This manual will adopt a "how to" approach covering relevant topics, including how to identify the broad well performance category, how to use and apply production type-curves to select remediation candidates, and how to use neural networks to identify root causes of well underperformance. In addition to the manual, ARI will provide copies of the type-curve and neural network software with a one-year, no-cost license along with the manuals. This project is scheduled to be completed early in 2002.

For further information on these projects, contact Gary Covatch at DOE NETL, phone 304-285-4589 or email Gary.Covatch@netl.doe.gov

Top of Page

Participants at DOE Workshop Reaffirm NPC Findings on U.S. Gas Supply

In a 1999 report, the National Petroleum Council (NPC) addressed "Meeting the Challenges of the Nation's Growing Natural Gas Demand." Recognizing the dynamics and changes within the natural gas industry, DOE coordinated a Washington workshop, held on March 5-6, to reassess the NPC's findings. More than 50 individuals, representing a wide array of energy interests including suppliers and transporters, electric utilities, trade associations and federal agencies, participated. Discussions focused on three areas: demand, supply, and transmission and distribution. Factors identified by workshop participants as warranting further examination or bench marking include, among others: (1) capability for fuel switching, (2) access to resources, (3) observed lag in production despite increased drilling, (4) Canadian natural gas supply, and (5) future costs for expanding gas transmission and distribution infrastructure. When full information from 2000 data become available, ideally in fall 2001, participants agreed that a later follow-on meeting would be beneficial.

DOE Contact: Nancy Johnson, phone 202-586-6458, email Nancy.Johnson@hq.doe.gov. Full report on workshop available at http://www.fossil.energy.gov/oil_gas/reports/gasworkshop/.