Abstract

Study of two-dimensional nuclear magnetic resonance (NMR) longitudinal and transverse relaxation time (T1–T2) responses of tight reservoirs and optimization of T1–T2 acquisition parameter are of significance for exploration and development of tight oil and gas reservoirs. In this paper, a tight sandstone digital model was generated using a micro-CT scanning technique, and the random-walk method was modified to simulate the T1–T2 measurements in a two-phase fluid-bearing tight sandstone. The echo data of the saturation-recovery Carr–Purcell–Meiboom–Gill pulse sequence were acquired and processed to obtain a T1–T2 map. We compared oil–water-bearing tight sandstone T1–T2 maps with gas–water-bearing tight sandstone T1–T2 maps, and analyzed how fluid types, various degrees of fluid saturation, acquisition parameters, and signal-to-noise ratio (SNR) of NMR data affected T1–T2 maps. The research indicated that NMR T1–T2 logging is more suitable for identifying gas-bearing reservoirs, rather than oil-bearing reservoirs. NMR-logging acquisition parameters have different influences on various fluid-bearing tight sandstone T1–T2 maps. Increasing either the magnetic field gradient or echo spacing leads to a leftward shift of the gas signal, but does not affect the water signal and the oil signal in a T1–T2 map; the T1–T2 map with a mixed group of wait time is better than that with a long or short group of wait time in providing fluid relaxation information. The NMR data with lower SNR result in stronger divergence of fluid signal in T1–T2 map, rendering it more difficult to identify the fluid according to the map, and a data SNR more than 21 is recommended.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call