Deepwater reservoirs often consist of highly laminated sand-shale sequences, where the formation layers are too thin to be resolved by conventional logging tools. To better estimate net sand and hydrocarbon volume in place, one may need to leverage the high resolutions offered by borehole image logs. Traditionally, explicit sand counting in thin beds has been done by applying a user-specified cutoff on a 1D resistivity curve extracted from electrical borehole images. These workflows require multiple preprocessing steps and log calibration, and the results are often highly sensitive to the cutoff selection, especially in high-salinity environments.
This paper presents a new method that estimates sand fractions directly from electrical borehole images without extracting an image resistivity curve or applying any preselected cutoffs. The processing is based on an artificial neural network, which takes the 2D borehole image array as input, and predicts sand fractions with the measurements from all button electrodes. A cumulative sand count can be computed after processing the borehole image logs along an entire well by summing up the estimated net sands. The neural network is trained and tested on a large dataset from wells in a deepwater reservoir with various degrees of laminations, and validated with sand fractions identified from core photos. Upon testing, a good match has been observed between the prediction and the target output. The results were also compared against another sand-counting method based on texture analysis, and showed advantages of yielding unbiased estimations and a lower margin of error.