No abstract

No abstract

The Junggar Basin is one of the main oil basins in Northwest China, and the northwest part of the basin is a significant petroleum exploration target. Mesozoic structural activity had an important influence on the present structural characteristics of the north-west Junggar Basin. In this paper, we discuss the tectonic evolution of the Junggar Basin by studying the structural characteristics of the Ke-Bai Fault. Our results suggest that the Ke-Bai Fault has obvious strike-slip characteristics and that it is an important basin-controlling fault that has controlled the tectonic evolution of the north-west Junggar Basin. The Junggar Basin was a rift basin in the Early Permian, but it experienced a tectonic reversal during the Middle Permian to the Triassic. The Ke-Bai Fault changed from a normal fault into a thrust fault from the Early Permian to the Triassic. During the Jurassic, dextral strike-slip structures developed along the Ke-Bai Fault, horsetail splay faults developed in its north segment, and there is evidence that it was under WNW-ESE compression. The dextral strike-slip faulting of the Ke-Bai Fault is consistent with tectonic movement along the Kelameili, Sangequan, Wu-Xia, Hong-Che, and North Tianshan fault systems in the Junggar Basin. We therefore infer that counterclockwise rotation occurred in the Junggar Basin in the Jurassic.

Manual fracture identification methods based on cores and image logging pseudopictures are limited by the expense and the amount of data. In this paper, we propose an integrated workflow, which takes the fracture identification as an end-to-end project, to combine the boundary detection and the deep learning classification to recognize fractured zones with accurate locations and reasonable thickness. We first apply the discrete wavelet transform algorithm and a boundary detection method named changing point detection to enhance the fracture sensibility of acoustic logs and segment the whole logging interval into non-overlapping subsections by estimating boundaries. The deep neural network based auto-encoders and the convolutional neural network classifier are then implemented to extract the hidden information from logs and categorize the subsections as the fractured or non-fractured zones. To validate the feasibility of this workflow, we apply it to the logging data from a real well. Compare with the benchmarks provided by the support vector machine , random forest and Adaboost model, the one-dimensional well profile predicted by the proposed changing point detection-deep learning classifier is more consistent with the manual identification result.