High-efficiency and high-quality detection of oil pipeline will significantly reduce environmental pollution and economic loss, so an unconventional oil pipeline anomaly detection convolutional neural network (CNN) algorithm based on attention mechanism is proposed in this article. By taking the simulated ground-penetrating radar (GPR) data as prior knowledge, the structure of the convolutional neural network based on the attention mechanism is constructed, and finally, the location and working condition of the underground oil pipeline are recognized in the simulation data and measured data. The simulation results show that after using the new optimized convolutional neural network, the accuracy rates of the leakage discrimination from horizontal data acquired along the oil pipeline and the classification of the target from longitudinal data acquired perpendicular to the oil pipeline are 94.5% and 84.6%, respectively. Compared with the original convolutional neural network without an attention mechanism, the accuracy rates of the leakage discrimination and the classification of the target are improved by 6.2% and 7.8%, respectively. We further train measured data with an optimized convolutional neural network, results show that compared with a conventional network, the new network can increase the corresponding accuracy rates of the leakage discrimination and the targets classification by 5.4% and 6.9%, reaching 92.3% and 84.4%, respectively. According to our study, the ground-penetrating radar oil pipeline recognition algorithm based on an attention mechanism can well accomplish the identification of underground oil pipelines.
Crosswell electromagnetic (EM) methods are widely used in subsurface geophysical prospecting because they can achieve more effective long-distance detection than single-well methods. However, a large-diameter borehole is required to increase the magnetic moment of the magnetic dipole source. For the long-distance detection of copper ores, which is usually performed in slim holes, we present a borehole-surface currentinjection-based crosswell EM logging method. Considering the cost of deploying casing, we inject a low-frequency AC directly into the ground, and converging current is formed around lowresistance anomalies in the formation. Then, the distribution of the anomalies can be inferred by detecting the low-frequency alternating magnetic field of the converging current in the receiver well. Moreover, to further improve the detection performance, we design a placement scheme for the grounding electrode for multianomaly crosswell detection based on the Gauss-Newton inversion algorithm, where the EM responses for different grounding electrode locations are analyzed. Field experiments are conducted using two slim open holes spaced approximately 1000 m apart for the detection of two copper ores. Through the processing and interpretation of measured EM signals, the conductivity imaging results of the crosswell EM method indicate that the measured distribution of anomalies is consistent with prior knowledge obtained from numerous single-well loggings, demonstrating the feasibility of the proposed application for long-distance crosswell EM logging in slim open holes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.