A context-aware system architecture for leak point detection in the large-scale petrochemical industry

In the large-scale petrochemical industry, one of the most concerning problems is the leakage of toxic gas. To solve this problem, it is necessary to locate the leak points and feed the possible location of leak points back to rescuers. Although some researchers have previously presented several methods to locate leak points, they ignored the impact of external factors, such as wind, and internal factors, such as the internal pressure of equipment, on the accurate detection of leak points. Fundamentally, both of those factors belong to context-aware data in a context-aware system. Therefore, this article proposes a context-aware system architecture for leak point detection in the large-scale petrochemical industry. In this three-layer architecture, a distributed database based on data categorization is designed in the storage layer, which is able to choose the most efficient approach to store the context-aware data from the gathering layer according to different context-aware data types. Then a real-time template matching algorithm for context-aware systems is presented in the computing layer to process the context-aware data stream. The architecture is a new scheme for accurate leak point detection, which is more consistent with practical application in the large scale petrochemical industry.