Tight oil is more and more important in petroleum industry. However, there is few systematic studies up to now. Based on core data, petrophysical experimental data, well logs and test data, a systematic study on seven-property relation of tight oil reservoirs is conducted in Lucaogou Formation, Permian, Jimusaer Sag, Junggar Basin, China. To be specific, firstly, lithology is identified via cross plot of matrix density and constructed structure index from nuclear magnetic resonance (NMR) log. Secondly, using iteration method, lower limit T2 relaxation time value of effective pore and oil-filled porosity are calculated to be 1.7 ms and 7 ms respectively, which is a new way to predict porosity and oil saturation in tight oil reservoirs. Thirdly, high resolution resistivity (Rt) curve is acquired from micro-resistivity imager log by dimension reduction process and piecewise calibration. Total organic carbon (TOC) content is predicted by curve overlapping method of interval transit time (AC) and the obtained high-resolution resistivity. Fourthly, according to the transformation of dynamic and static rock mechanics parameters, an improved method is proposed for quantitatively predicting brittleness index (BI) of rocks via well logs. Ground stress value and direction are evaluated by various methods comprehensively. This study obtains a good application effect and provides important technical support for evaluating tight oil reservoirs via well logs, which plays an important role in tight oil exploration and development and new oil field discovery.

Keywords:
well logging, machine learning, Upstream Oil & Gas, porosity, oil saturation, reservoir, Artificial Intelligence, study area, formation evaluation, seven-property relation
Subjects:
Formation Evaluation & Management, Open hole/cased hole log analysis
This content is only available via PDF.
2019. International Petroleum Technology Conference
You can access this article if you purchase or spend a download.