From the President

Dick Merkel
EnCana

First and foremost, we need to thank Fronterra Geosciences for sponsoring the beer at last month's luncheon.  Our sincere thanks were inadvertently forgotten at the meeting.

Last month Steve Cumella shared his experience and showed some pitfalls of using pulsed neutron logs in tight gas sandstone reservoirs (Piceance Basin). His knowledge of tool responses and processing techniques from multiple venders is invaluable.

One of the log processing techniques used to evaluate pulsed neutron data incorporates the use of neural networks. Neural-nets are typically used to either reconstruct synthetic open hole data or to directly determine hydrocarbon volumetrics (by open hole log interpretation model emulation).

This month, John Quirein will discuss the theory and applications of neural-net models, and how Halliburton is applying this technology in their interpretation package. For those who want a head start on John's talk, please refer to paper R in the 2003 SPWLA Transactions for a paper he published on neural network techniques.

As you probably know, Vicki King, SPWLA's Executive Director, is suffering from a very advanced and aggressive form of breast cancer. On behalf of the DWLS, your Board of Directors sent her a planter of flowers. She was deeply touched at receiving this and wanted me to thank each of the members of DWLS.

From the Vice President - Membership

Bill Rodgers
Weatherford International

We have added 82 new members so far this year. This is a total of 216 active members for the 2007-2008 year. We continue to grow!

I would like to thank all of our members that have encouraged all geoscience professionals who are interested in well logs and petrophysics to become members in the Denver Well Log Society. I think our guests speakers have given excellent presentations this year have been instrumental in generating interest in the DWLS.

Please continue to invite all geoscience professionals to join the DWLS.

For membership, please download the membership form and send with your payment to me at:

Bill Rodgers
DWLS VP Membership
c/o Weatherford International
410 17th St, Suite 400
Denver, CO 80202
mailto:bill.rodgers@weatherford.com

Or bring your filled out form along with payment to the February meeting.

If you are already a lifetime member, you may want to fill out an update form if any of your contact information may have changed in the last year.

From the Secretary / Treasurer

Stefani Whittaker
The Discovery Group

The DWLS Board of Directors has recently enacted a new no-show policy for the luncheon meetings

"As of February 1, 2008, DWLS will be enforcing a policy for a member who signs up for the monthly luncheon but does not attend. A member must cancel no later than 10am on the day prior to the luncheon, or send a paying replacement. This stand-in must inform the person collecting money whose reservation they will be filling. If a member fails to cancel or send a replacement they will carry a balance on their account to be paid at the next luncheon attended. In order to be fair to all, DWLS reserves the right to sell a lunch if a member does not arrive by the time the first course is served at 11:45am. Although DWLS may be able to sell an unclaimed lunch to a walk-in, this does not exempt the member from paying $20 for the reserved lunch. To cancel a reservation, please contact Eleice Wickham at (303)-573-2781 no later than 10am on the day prior to the luncheon."

If you are interested, you may find a copy of our current year-to-date P&L statement here.

Finally, I still have 10 copies of the Fall Workshop course notes if anyone's interested in purchasing them.

"Chi Modeling: Predicting Open-Hole Logs From Cased-Hole Logs"

John A. Quirein and Marshall Jung
Halliburton

Presented is a modeling system (CHI Modeling System) that generates accurate, synthetic resistivity, neutron porosity, and density ("triple combo") open-hole logs from cased-hole pulsed neutron capture (PNC) logs. This technology can eliminate the requirement for running open-hole logs in most new wells in many fields and can also provide synthetic open-hole logs, including magnetic imaging logs, in almost all older cased wells. Some benefits that result are savings in rig time associated with running logs in uncased holes and reduced risk of stuck tools containing radioactive sources. This capability greatly facilitates cost effective acquisition of logs and their correlation in both green-field and brown-field applications.
 
The CHI modeling system utilizes previously-run open-hole triple combo logs and cased-hole pulsed neutron logs from the same well to form "training wells".  Using these training wells together with genetic algorithms a nonlinear neural network model is developed to establish direct mapping between selected PNC inputs and the open-hole log curves. These models are then utilized to process cased-hole PNC measurements from different wells in the field having no open-hole logs. The PNC data is input to a novel optimized ensemble of neural networks (NNE), since the complexity of logging situations is such that single neural network solutions are often unreliable. The NNE generates the simulated open-hole logs with accuracy unachievable in prior log simulation technologies. The process is automated for a field once the NNE model is developed.  An assortment of field examples is presented to demonstrate the success of the system.

About the Speaker

John A Quirein is currently a Senior Technical Advisor for Halliburton Energy Services in the Petrophysical Development Group, focusing in the areas of interpretation and software development. Prior to that, he worked 12 years for Mobil in the areas of petrophysics, geology, and geophysics where he directed the development of Mobil's in-house petrophysical software program. He co-developed the Mobil regression and neural network based approach for predicting rock properties from seismic, by applying seismic attributes calibrated to well-log data. Prior to that, he worked 10 years for Schlumberger where he developed interpretation algorithms for the LDT, NGT, and GLT logging tools and was the lead engineer for the development of the ELAN multi-tool, multi-model software interpretation program.
 
Marshall Jung is currently a Senior Log Analyst with Halliburton in Denver.  He specializes in cased-hole solutions and interpretations.  Before coming to Denver he worked as an engineer for Halliburton primarily in the Piceance Basin.  He holds degrees in Mathematics and Economics from Colorado school of Mines.