Ultra-short-term Wind Power Prediction Of Large-scale Wind Farm Considering Wind Speed Information

In recent years,emerging clean energy is increasingly booming,among which,wind power has performed particularly well.Wind energy is mainly derived from near-earth winds in the natural atmosphere,with obvious volatility and intermittent.How to use wind energy effectively,wind power predicting is considered as an important research topic.In the current research background of wind power predicting,this article makes a decision analysis on wind power data and further improves the accuracy of wind power ultra-short-term prediction.There are many types of wind power data such as wind speed,wind direction,power,pitch angle and other elements.This article mainly discusses wind speed information and power data,this kind of data usually contains a massive system disturbances,hence it has a sharp impact on the prediction results and needs to be analyzed and processed before prediction.Through the separation method of high-frequency components and low-frequency components,it can identify and eliminate high-frequency components in the wind power data system,and the impact of high-frequency components on the stability of prediction model system is reduced,then improve the accuracy of prediction.At the same time,quantitative correlation analysis is conducted on wind speed and power sequence,and a decision model of wind power data information based on grey correlation decision is established.In order to select the appropriate decision variables,the probability density analysis process based on grey correlation is introduced to obtain the quantitative indicators that describe the relationship between wind speed fluctuations and power fluctuations,which is input to prediction model for calculation as a weight variable.This article mainly discusses the basic establishment principle of the prediction model and the target input and output data,and describes the principle of the pre-predicted data processing.It can be seen from the decision-making that the degree of freedom can be used as a quantitative index to describe the relationship between wind speed and power of the wind farm.And on this basis,two kinds of ultra-short-term prediction method for wind power of large scale wind farms with wind speed information are proposed.They are Real-time prediction of wind power based on probability distribution and gray relational decision-making considered time series,and Real-time prediction of large scale wind farm space rise scale based on fractal scaling factor transform considered the fusion of time and space dimensions,the prediction results and analysis of measured data show that the average of two predict methods the accuracy are improved.In order to make the comparison,and two kinds of traditional wind power ultra-short-term prediction model without wind speed information are put forward,they are: Real-time prediction for wind power based on Kalman Filter and Suport Vector Mahines and Wind power real-time prediction based on Grey Buffer Operator–Kalman Filtering.The simulation proves that the introduction of wind speed information can not only express the wind to power truly,but also improve the prediction accuracy of ultra-short-term wind power in large-scale wind farms accurately.