<![CDATA[The oil industry plays a crucial role in Iraq's economy. There's a growing need for technologies that can quickly detect leaks in oil pipelines because leaks can have serious ramifications, including monetary losses, endangerment to public safety, environmental degradation, and resource waste. Advances in technology and software have made it possible to detect leaks. Current approaches often require manual extraction of features, which can be slow and inefficient. This paper presents a new method that proposes using convolutional neural networks (CNNs) for automatic feature extraction. The Iraqi Ministry of Oil, specifically the Basra Oil Company, provided the dataset, such as total distance (km), pressure (bar), and flow rate (STB/d). We split the data into training (70%) and testing (30%) sets. then we calculate metrics such as confusion matrices, accuracy, precision, recall, and F-score to evaluate performance and calculate errors from the regression analysis (root mean square error, root mean absolute error, and relative error). Our contribution to this work is to use 1DCNN to identify leaks, pinpoint their location, and even predict the amount of spilled oil, unlike other research that only uses it to evaluate the presence or absence of a leak only. Additionally, we've created a user-friendly interface for the system. Finally, compare the proposed approach with conventional and alternative methods to show its efficiency. In the future, we plan to expand the system to assess pipeline corrosion and predict its remaining lifespan.]]>
Copyrights © 2024
