Research And Application Of Key Technologies Of Intelligent Wind Power Early Warning Platform

The air pollution caused by traditional thermal power generation is no longer consistent with the concept of sustainable development.Among them,the wind power generation relies on the wind turbine to convert the wind energy into the electricity,does not need to consume the fuel,the environmental protection clean,is more concerned by the industry.The key to wind power generation is to ensure the stable output of the fan and to do a good job in the daily maintenance and monitoring of the fan.However,in the actual construction,the fans are usually set up in a wide and sparsely populated area with remote location,which makes the fan inspection and maintenance difficult and costly.To effectively solve this problem is the key to the development and stability of power generation enterprises,it is also an important issue in academic research.Based on this,this paper developed a smart wind power early warning platform.The system provides visualized on-line functions such as real-time information monitoring of the fan,historical information inquiry,and early warning information of the fan,which replaces the working mode of manual inspection,greatly improves the work efficiency,saves considerable economic cost,and plays an important role in preventing the fan failure monitoring and maintenance.Paper with a wind power company wind warning is based on the actual project of platform construction,according to the development of software engineering method,on the basis of demand analysis,by adopting the idea of before and after the separation,unified data interface on the system design and development,scheduling,early warning information system mainly includes the data management,early warning model,data storage and data interface services five function module.This paper focuses on the early-warning platform architecture of Angular and Spring Boot,establishes the data service interface,and realizes the fast and stable interaction of a large amount of data between the front and rear ends.Based on the study of the fan fault warning mechanism,a fan fault warning model based on LSTM is proposed,and the fan fault warning system is implemented under the framework of Tensor Flow,which improves the accuracy of the fan fault warning compared with the traditional model.The system has been running steadily for about a year since its launch,with continuous real-time monitoring and high warning information accuracy,achieving the expected fan monitoring and warning effect.The main innovations of this paper include the following aspects :(1)combining the Angular architecture and the Spring Boot architecture to design the early-warning platform architecture;(2)designed data scheduling algorithm and early warning model rule configuration algorithm;(3)proposed a fault warning model based on LSTM.