Inversion-based petrophysical interpretation of logging-while-drilling nuclear and resistivity measurements

Abstract

Interpretation of borehole measurements acquired in high-angle (HA) and horizontal (HZ) wells is challenging due to the significant influence of well trajectory and bed geometrical effects. Experience shows that accurate integrated interpretation of well logs acquired in HA/HZ wells requires explicit consideration of 3D measurement physics. The most reliable alternative for interpretation of well logs in HA/HZ wells is with inversion techniques that correct measurements for shoulder-bed, undulating well trajectory, and bed geometrical effects while taking advantage of high data resolution. We discovered an efficient layer-based inversion workflow for combined, quantitative petrophysical and compositional interpretation of logging-while-drilling sector-based nuclear (density, neutron porosity, photoelectric factor, gamma ray) and array propagation resistivity measurements acquired in HA/HZ wells. A challenging synthetic benchmark example confirmed improved formation evaluation with the layer-based inversion workflow across hydrocarbon-bearing zones in HA/HZ wells, where estimated hydrocarbon pore volume and porosity increased by 10% and 15%, respectively, with respect to conventional interpretation methods. Furthermore, application of the inversion-based method to a field example of HZ well across calcite-cemented siltstone layers confirmed its advantage over conventional interpretation techniques.

2D inversion-based interpretation of logging-while-drilling thermal-neutron and gamma-ray time decays

“…Mendoza et al (2010) correct nuclear porosity measurements for shoulder-bed and geometric effects in deviated wells using a 3D fast-forward algorithm and linear inversion to estimate layer-bylayer porosities. Similarly, Ijasan et al (2013) develop a fast-inversion algorithm for nuclear porosity and LWD propagation measurements that simultaneously accounted for environmental and geometric effects for both measurements, which led to improved estimations of hydrocarbon pore volume.…”

A deep learning approach to the inversion of borehole resistivity measurements

“…One of the purposes of geophysical measurements is to interrogate the subsurface of the Earth to find oil and gas, and to optimize the production of existing hydrocarbon reservoirs. We divide existing geophysical measurements into two categories: (a) surface geophysical measurements, such as controlled source electromagnetics (CSEM) (see, e.g., [1,2]), seismic (see, e.g., [3]), and magnetotellurics (MT) (see, e.g., [4]), and (b) borehole sensing, such as loggingwhile-drilling (LWD) data (see, e.g., [5,6]).…”

Adjoint-based formulation for computing derivatives with respect to bed boundary positions in resistivity geophysics

“…We consider resistivity measurements to characterize the electrical properties of the subsurface. There exist: (a) on surface resistivity measurements acquisition systems such as controlled source electromagnetic (CSEM) [1][2][3][4] and Magnetotellurics (MT) [5,6], and (b) borehole logging measurements such as those obtained with logging-while-drilling (LWD) devices [7][8][9], including the so-called deep and extra-deep [10,11] logging devices used for geosteering purposes. Recently developed LWD resistivity measurements are able to measure all nine components of the magnetic field, namely H xx , H xy , H xz , H yx , H yy , H yz , H zx , H zy , H zz , where the first and second sub-indexes indicate the orientation of the transmitter and the receiver, respectively.…”

“…Resistivity measurements are inverted to map the Earth's subsurface (see, e.g., [3,7,8]). Using gradient-based inversion techniques (e.g., Gauss-Newton), we need to estimate the derivatives of the simulated measurements with respect to the inversion variables to form the Jacobian matrix.…”