Big Data Analytic for Pipeline Monitoring

点击翻译按钮获取中文摘要

In recent years, the Canadian public has shown an increased level of sensitivity to the environmental impacts of pipeline leaks as was the case with the recent Nexen pipeline leak, which was the largest leak in Alberta history. A pipeline leak or spill can be costly, dangerous, and time-consuming. It is important to be able to sense and detect the leaks when they are small. On the other hand, with improved sensor technologies, massive amount of data are generated by these pipeline monitoring sensors. Hifi Engineering Inc. is a Canadian company specializing in developing robust fiber optic sensing technology for monitoring fluid flow in pipelines, wells, and critical infrastructure. Their goal is to increase reliability and safety of wells and pipelines by providing a distributed dynamic sensing system to monitor integrity and detect leaks. To be able to detect small leaks based on the large amount of data, we will work with Hifi Engineering Inc. to explore the use of big data analytics to improve leak detection with massive data and to develop a model of the fluid flow in the pipeline using the fiber optic sensing data. The project tries to improve the size of the data to be analyzed, while quickly and accurately detecting disturbances. The sensor data (acoustic, temperature, strain data) provided by Hifi Engineering will be used as the dataset to analyze for leak detection. The environmental benefits of this project will also be important as a number of high profile pipeline projects, with the potential to create many jobs, are currently on-hold, partially due to the unavailability of reliable leak detection technologies.