LoRa Network Enhancement of Oil Pipelines Leak Detection System Based on Artificial Intelligence

Detection of minor leaks in oil and gas pipelines is a critical and persistent problem in the oil and gas industry. Many organizations have long relied on fixed hardware or manual assessments to monitor leaks. With rapid industrialization and technological advancements, innovative engineering technologies that are cost-effective, faster, and easier to implement efficient oil pipelines leak detection system. In this context, a Wireless Sensor Network (WSN) is increasingly required to optimize the reliability of the inspection and improve the accuracy of the real time oil pipelines leak detection and localization monitoring system. A LoRa WAN solution is proposed in this paper to detecting and locating leaks for long distance oil pipelines. This research will focus on applying several machine learning techniques to enhance the network parameters such as power consumption, packet loss and delay in order to maximize the durability of individual nodes. The system implementation will involve gathering data from <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">OMNeT++ network</tex> simulator which integrated with python script and several libraries and frameworks such as Flora, Inet and tensor flow in order to run machine learning over the Flora network. The hybrid machine learning algorithm, which consists from, random forest tree, support vector machine, k-mean clustering algorithms were used with an accuracy 89% to develop the model and enhance the LoRa network parameters such as power consumption, packet delay and packet loss and reached as example for power consumption to enhancement ratio around 48% when compared with default LoRa network.