Research On Monitoring Data Management For Offshore Platforms Based On Internet Of Things

Offshore platforms are large industrial facilities for offshore drilling operations. The service environment of the platforms is complex and changeable. It is difficult to simulate the external excitations in the laboratory. As a result, the environmental loads and structural responses data need to be measured through the long-term prototype monitoring and these data can provide important reference for analyzing the health status of the platforms. A large amount of data have been collected during the long-term prototype monitoring, because the monitoring conditions are special and the monitoring data are large-scale, dispersal sources, fast produced and heterogeneous, the traditional management method cannot satisfy the application requirements. So efficiently managing, searching and analyzing these data becomes the key problem in the offshore platform monitoring. Aimed at the problem of long-term prototype monitoring data management for "Tiao Zhan Hao" platform in the South China Sea and multiple FPSO in the Bohai Sea, a monitoring data management scheme based on internet of things is proposed in this paper. According to this scheme, a corresponding data management system is designed, which can solve the problems of data transmission under the non-stationary internet environment, automatically normalized importation of the heterogeneous data flow and efficient storage and retrieve of the monitoring data.The offshore platform monitoring requires high demands of the real-time performance of the data collection, but the internet condition in the ocean environment is not stable and a lot of accumulated data need to be transported when the system is recovering from the interrupt under the unattended condition, so the method of artificial transmission and remote operation are not suitable. To solve these problems, a real-time and unstable internet-oriented data transport protocol is introduced. Under the condition of existing communications, the monitoring data are automatically collected from multiple platforms using this protocol and it can reduce manual operation, realize the automatic data transmission and solve the problems of automatic data collection of the data management system under the unattended and broken internet conditions. During the long-term monitoring of the offshore platforms, different monitoring variables, sensors, ways of data collection and the update replace of the software and hardware make different forms of data stream difficult to be transformed into the monitoring database, the data storage process needs manual operation. Therefore, an automatic method of data stream importation is presented. The monitoring data stream is identified and normalized according to the semantics of the data and format. The data collected from multiple systems are stored into the database automatically, which solves the problems of automatically normalized data and importation using multiple sensors for heterogeneous data stream.The monitoring data collected during the long-term offshore platform monitoring are correlative and in different data structure and format, traditional file management system and data management system cannot satisfy the storage need for the large-scale and long-term monitoring and retrieve need for the data analysis. Besides, the monitoring data are in the format of time series, frequency and modes, which cannot be suitably stored in traditional way. In this paper, a kind of hybrid storage architecture is proposed to solve the efficient management of data in different formats and it can provide efficient support for high speed data search, complex inquire and statistic analysis that are data-oriented.