Offshore Wind Power Safety Early Warning and Dynamic Visualization Platform Driven by Multi-Source Heterogeneous Data

With the advancement of the “dual carbon goals” and the surge in demand for deep-sea wind power development, there are challenges for offshore wind power such as difficulty in integrating multi-source heterogeneous data, delayed safety warnings, and low efficiency of dynamic visualization. This study constructs an offshore wind power safety warning and dynamic visualization platform based on Service-Oriented Architecture (SOA) structure, integrating Hadoop distributed system and multi-source data integration technology to achieve efficient management and service of data. We adopt a hybrid database architecture of HDFS, HBase and PostgreSQL, which improves storage efficiency by multi-dimensional data dimension reduction and row key optimization strategies, combines Spark memory computing framework to achieve billion-level data statistics, with an efficiency improvement of 41%, and integrates particle system and WebGL technology to break through the bottleneck of dynamic visualization of marine environment, improving efficiency by 130 times compared with Canvas2D rendering. The results show that the platform supports multi-dimensional parameter visualization of wind energy resource assessment module, the safety warning module realizes 72 -hour typhoon path prediction, and the response efficiency of intelligent operation and maintenance module is significantly improved, effectively solving data islands and delayed warnings. The platform provides a standardized technical solution for offshore wind power safety management, promotes the intelligent upgrading of wind power industry and the optimization of energy structure, and has practical significance for achieving the large-scale development of deep-sea wind power and dual carbon goals of China.