Monitoring and Controlling Oil and Gas Drilling Platforms Based on Unmanned Aerial Vehicles and Remote Sensing Technology
Yu Hu · 2024
Large-scale monitoring operations cannot be finished quickly by traditional oil and gas drilling (OGD) monitoring since it depends on physical labor. The utilization of remote sensing (RS) and unmanned aerial vehicle (UAV) technologies can enhance the precision of monitoring and real-time performance of OGD platforms. Preprocessing was carried out on the large images of OGD platforms that were acquired by using RS and UAVs in order to ensure image quality. Improvements were developed based on the Residual Network (ResNet) model, such as the addition of multi-scale convolution, the attention mechanism, and the utilization of global max pooling for timely monitoring of anomalous OGD scenarios. According to the experimental findings, the enhanced ResNet model outperformed ResNet, SVM, RF, and Naive Bayes in terms of accuracy, with an average accuracy of 97.0% for OGD fault classification. As a result, OGD platforms may be fully monitored and controlled by the use of UAVs and RS technology, which can also precisely identify anomalous OGD failures.