Borehole coring is costly and provides discontinuous information. How can electrical property variations of the borehole wall and adjacent rock formations be obtained continuously? Electrical borehole logging is a key technology for solving this problem. Its device consists of current electrodes A and B and measuring electrodes M and N, which form an electrode array. Apparent resistivity is calculated by the formula Ra = K·ΔU/I. However, apparent resistivity is affected by borehole diameter, mud resistivity, and surrounding rock, and is not equal to the true formation resistivity; it therefore requires correction and inversion.
Measurement Methods: From Conventional to Imaging
Building on conventional apparent resistivity logging, micro-resistivity scanning imaging logging (MRSI) uses densely arranged button electrodes on a pad (spaced 2.5 mm apart) to convert borehole-wall resistivity variations into visual images. Research in 2025 shows that a load-adaptive transmitter module extends the applicable range of drilling fluid resistivity to three times its original value and increases load adaptability to four times, significantly enhancing fracture and thin-bed identification in high-resistivity drilling fluids and high-resistivity formations. Progress has also been made in multi-parameter determination methods for micro-resistivity scanning imaging while drilling (MRSI-LWD) in oil-based drilling fluids. By combining numerical modeling with a BP neural network, formation resistivity, standoff distance, and high-frequency dielectric constant can be obtained simultaneously, with a coefficient of determination exceeding 99%.
Multi-Parameter Fusion and Joint Observation
The borehole induced polarization method measures polarizability parameters and can determine the spatial distribution of ore bodies in sulfide-type metal ore exploration. Azimuthal induced polarization logging uses differences in the shapes of induced polarization anomaly curves at different azimuths to determine the azimuth, distance, and burial depth of an ore body relative to the borehole. Joint numerical simulation of borehole-to-surface DC resistivity and time-domain induced polarization uses differences in polarization characteristics to distinguish anomalous bodies, improving the reliability of fluid identification in oil and gas field development.
Intelligent Inversion and Application Expansion
Artificial intelligence is accelerating its integration into electrical borehole logging data processing. After training on 1,000 samples, a random forest method can accurately predict pile foundation burial depth and has strong robustness to noise. A semi-supervised learning inversion method based on forward modeling constraints, trained with both labeled and unlabeled data, effectively improves the accuracy of high-density electrical inversion and its ability to characterize complex geological structures. A dual-optimization model of “genetic algorithm + neural network” has been applied to a 1D electrical inversion software system, promoting intelligent geophysical data processing. Electrical borehole logging has evolved from single apparent resistivity measurement into a technical system advancing in three directions: high-resolution imaging, multi-parameter joint observation, and intelligent inversion. Its applications cover deep metal mineral exploration, oil and gas reservoir evaluation, and urban engineering investigation. In the face of complex borehole conditions and deep targets, how to improve inversion stability, reduce non-uniqueness, and achieve real-time while-drilling interpretation remains the key issue for the next stage. For technical engineers, method selection requires a comprehensive trade-off among resolution, depth of investigation, borehole adaptability, and cost.
