Wind Farm Operation Cognition And Prediction Based On Big Data And Machine Learning Method

With the increasing penetration of wind power in the power system in recent years,wind power plays an indispensable role.The massive data generated during the operation of the wind turbine records the actual operation of the wind turbine.Therefore,it is necessary to carry out analysis and extract valuable information by adapting big data analysis and machine learning.With the rapid development of big data and machine learning technology,wind power big data research has greater possibilities and broader application prospects.To reseach on the application of massive data generated during the operation of wind farms,this paper explores the effective information inside the wind farm big data along the order of big data preprocessing,wind farm operation status visualization,wind farm operation data analysis and forecasting.Firstly,the basic composition,development status and application prospects of wind power data are introduced.The application of big data research methods and machine learning in wind farm are analyzed.Secondly,the big data preprocessing methods are introduced.Based on the measured data,principal component analysis and t-distribution neighborhood embedding algorithm are applied to reduce the dimensionality of high-dimensional data collected during wind farm operation.Thirdly,the visualization system for statistical analysis and real-time status monitoring of wind power big data is designed and implemented.The operation status of wind farm is realized by combining the time series database InfluxDB and the visualization tool Grafana.Finally,the wind turbine operating state model based on long short-term memory network is built and related factors are predicted.The Prony algorithm is applied to analyze the oscillation module and explore the interaction between grid and wind farms.