Assessing the Capabilities of a Cased-Hole Reservoir Analysis System in the Gulf of Thailand

Abstract

The option of using a cased-reservoir analysis system as an alternative to traditional open-hole log analysis was tested as a method to reduce drilling rig time. On a series of five wells in the Gulf of Thailand, an integrated pulsed-neutron system that measures Sigma, inelastic spectroscopy, neutron porosity and density was used to drive the cased-hole analysis. The capabilities and accuracy of the cased-hole results were compared to open-hole porosities and water saturations. One concern for cased-hole logging is accuracy in this well-completion environment. These wells are completed as monobores with 2 7/8-inch tubing cemented directly into the 6 1/8-inch open hole (no well casing). Initially, the pulsed-neutron porosity coefficients for 5 1/2-inch casing were used with reasonable accuracy. To increase the porosity accuracy, the data from this test served as input to an artificial neural network for determining porosity coefficients in this specific borehole environment. The project consisted of 12,099 feet of log data including 30 gas sands, 5 oil sands, and numerous water sands in the five wells. The results of this assessment are shown via log examples over specific intervals. In addition, several cross-plots comparing open-hole data to cased-holed data show the results of this evaluation. Graphs of net pay, net effective porosity, net shale volume, and water saturation are shown for the entire project. Introduction Figure 1 is a map of fields in the Gulf of Thailand. Production from the fields operated by Unocal is approximately 1 BCFD of natural gas1 and 33,000 BPD of condensate.2 Since these reservoirs are highly compartmentalized and have limited drainage, it is critical for well-construction costs to be minimized. One technique that is appropriate for this reservoir environment is to drill slim wellbores and complete as monobores with 2 7/8-inch tubing cemented in open hole (no well casing). To reduce drilling-rig time and lost tool liabilities, a test of the capabilities of a cased-reservoir analysis system was devised. Five wells logged with a standard open-hole "triple combo" were selected to run the Computalog PND™ pulsed-neutron system. In a single logging pass this system measures Sigma, C/O ratio, neutron porosity and density. In this assessment the cased-hole porosities were of prime importance, and the cased-hole gas and oil saturation measurements were of secondary importance. These reservoirs are predominantly non-marine fluvial sands with interbedded shales and occasional coals. The logged intervals consisted of two different geometries: the upper sections are 8 1/2-inch boreholes with 7-inch casing and 2 7/8-inch tubing (casing-borehole and tubing-casing annuli cemented), the lower sections are 6 1/8-inch boreholes with 2 7/8-inch tubing centralized and cemented. Reservoir fluids are oil, hydrocarbon gas with carbon dioxide, and fresh waters (>5Kppm NaCl). The study was restricted to the shallow part of the productive section by the maximum temperature specification for the PND™ of 300F. At the studied depths (true vertical depth from -3500 to -6190 feet) the reservoir sequences have a low percentage of carbon dioxide (typical >10%). The low salinity of the formation water means Sigma will have poor oil-to-water resolution. However, the low cross-section of gas can be exploited to make a gas saturation calculation. The Sigma measurement is fairly borehole- independent and standard water saturation evaluations were done using the variable matrix model.3 The Carbon/Oxygen saturation measurement is very borehole-dependent, and characterization for this wellbore environment is discussed in the Results: Saturation from C/O section. Characterization of the porosity measurements is discussed in the following section.