Research On Wind Power Forecasting And Wind Parameters Monitoring System For Wind Farm

Wind energy has strong randomness,intermittentity and uncertainty,which makes wind power possess great instability and bring various difficulties for wind power grid integration.Therefore,accurate forecasting of wind speed and wind power has great significance for maintaining the stability of power grid and promoting the further development of wind power industry.In order to improve the forecasting accuracy of wind power,this paper studies an improved GABP neural network wind speed forecasting model,and proposes a wind power forecasting model based on wind speed forecast correction.In order to achieve the collection of wind parameters and the assessment of wind resources,as well as provide supplementary data for wind speed forecasting and wind power forecasting,a multi-functional wind measurement parameters monitoring system is designed in this paper.The main research contents of this paper are as listed below:(1)The paper expounds background and significance of the subject,and summarizes the development trend of wind power.The typical wind power forecasting systems at home and abroad are introduced,and the forecasting methods are summarized which are frequently used for wind speed and wind power at present.This paper also analyzes the current research status of acquiring and monitoring wind parameters.(2)In order to improve the quality of wind data,this paper combines the specific situation of wind farms with the latest relevant studies according to the national standard GB/T18709-2002.The wind data are checked and handled in terms of rationality,relativity and variation tendency,etc.Furthermore,this paper adopts the actual measured data of a wind farm in Shanxi Province to analyze the wind distribution and variation properties.(3)The paper studies an improved GABP neural network wind speed forecasting model from three aspects of confirmation in input variable,choice of training samples based on clustering algorithm,and the proposed optimal weight and threshold of neural network.The establishment of neural network model is highly dependent on training data.In this paper,clustering algorithm is used to select training samples.And through the simulation study of actual data,the method of this paper is effective.(4)The paper proposes a wind power forecasting model based on wind speed forecast correction to reduce the forecasting error and improve the forecasting accuracy.The paper also compares and analyzes the wind power model based on power curve and the wind power model based on the improved GABP neural network.(5)The paper designs a multi-functional wind parameters monitoring system in the views of high reliability and strong extensibility.The system includes parameter detection layer,data acquisition layer,monitoring layer and so on.The system has two kinds of local and remote monitoring,friendly interaction in human-machine interface and more effective monitoring effect.