Industrial big data analytics practice based on data lake platform: an example of intelligent oilfield energy efficiency analysis
Li Man1, An Chuangfeng1, Gao Jing1, Niu Yongsheng1，Yao Jiakun2
1.CNOOC (China) Tianjin Branch; 2.China Electronics System Technology Co.
Abstract： As industrial systems and the application scenarios of industrial enterprises become increasingly complex, the quantity and variety of data processed by the system also increase. In light of the growing number of diverse business application scenarios and the increasing volume of heterogeneous data from a multitude of sources, the need for enhanced mobility and flexibility in data analysis is becoming increasingly apparent. The conventional database-centric approach to big data analysis is inadequate for accommodating the heterogeneous structural characteristics of data sinks and the evolving nature of data sources. Accordingly, this paper presents a comprehensive, integrated and collaborative big data analysis application implementation framework. This framework combines the data lake platform and intelligent algorithms to establish an analysis model for industrial big data. Furthermore, driven by business analysis requirements, the data lake platform′s processing, aggregation, and management capabilities are leveraged to efficiently carry out data modeling, preparation, testing, training, and validation. Subsequently, the application is verified in an intelligent oilfield energy efficiency analysis scenario. This successfully demonstrates the system′s ability to predict, optimize, and make decisions in this context, providing data support for the entire business process of oil fields and promoting the sustainable development of intelligent oilfields.
