Statistical Analysis And Inference Of Wind Turbine Deviation Detection And Application Based On New Energy Big Data Platform

The new energy big data platform can provide valuable online and offline integrated services for enterprise customers based on massive new energy big data and mature models and applications.With the promotion of industrial big data technology,data analysis model related to wind power in the big data platform are also increasingly applied to various wind farms.Excessive deviation of the wind turbine from the wind will lead to the reduction of power generation efficiency,and will also cause problems such as fan component wear.The wind monitoring assistant for wind turbine yaw is a specific industrial application of the platform and a practical case of big data analysis technology helping wind power enterprises.Through this application,the Wind farm station can timely find and correct the wind turbines that have problems with the wind angle.The effect of improving the power generation of the existing units is very obvious,and the input cost is low,which helps the wind power enterprises to reduce costs and increase efficiency.This thesis first puts forward the construction ideas of the new energy big data platform,analyzes the core resources and core business of the platform,and gives the main business functions of the data model and application in the big data platform.Then,business understanding was carried out for the problem of wind turbine yaw,the deviation correction scheme of wind turbines to wind yaw is proposed,and the data source,data understanding,algorithm principle,algorithm flow and effect evaluation method of monitoring model for wind deviation of wind turbine are given.The model algorithm includes data preprocessing,wind speed segmentation,wind speed smoothing,feature engineering,quantile regression analysis and non-linear regression analysis and other key technologies.Based on the relevant resources and services of the data analysis model pipeline provided by the new energy big data platform,the constructed monitoring model for wind deviation of wind turbine is transformed into an industrial application-the wind monitoring assistant for wind turbine yaw,and deployed to the big data platform.This application provides the monitoring alarm and automatic deviation correction function of wind turbine to wind yaw.It can access the wind turbine operation data collected by the front end equipment of the wind farm through the big data platform,automatically issue the deviation correction command,realizing the optimization and closed-loop of wind turbine to wind deviation.The platform-based application is provided to users of wind power enterprises.Through the new energy big data platform,the whole process exploration and closed-loop landing of data analysis models and applications such as business understanding,data display,data exploration,data preprocessing,model construction,model release,application deployment,application launch and use have been realized.Finally,summarize the research results,make supplements for the deficiencies,and propose further research directions.