Wind Power Range Estimation Based On Operational Condition Identification

As a mature and clean energy,wind power is widely used.However,the intermittent and fluctuating nature of wind power poses challenges to the operation of the power grid.The variability and complexity of wind turbine operating conditions also affect the accuracy of power prediction.There fore,the identification of operational conditions has become an important direction for wind power research.This paper first makes a deterministic prediction of wind power,then identifies the operating conditions of the wind turbine,finally completes t he interval prediction of wind power.The specific content is divided as follow:Firstly,in view of the problems of EMD modal aliasing and excessive IMF components,a deterministic forecasting method for wind power based on MEEMD and permutation entropy is proposed,which improves the accuracy of t he model compared to other methods and paves the way for subsequent interval prediction.Secondly,the operating characteristics of wind turbines are analyzed.In order to solve the problem of poor condition identification with power,SFCM-based wind turbine condition identification method is used.SFCM can reuse the sample data in the transition conditions,which solves the problem that the number of samples in certain conditions is small in the method of identi fying the conditions only with the real power,and it is better than the indicators in coverage and average bandwidth evaluation.The method for identifying operational conditions by real power improves the effectiveness of operational condition identification.In addition,the research on the interval prediction of wind power is carried out.Based on the kernel density estimation method,the probability density cur ve of the deterministic prediction error under each operating condition of the wind turbine is calculated,and the confidence interval of each operating condition is calculated according to statistics.The results are given in the form of,and the interval prediction is completed.Finally,based on the results of interval prediction,a method for amending the deterministic prediction model based on the results of interval prediction is proposed.The deterministic prediction model is further improved from th e perspective of interval prediction.Corrected to improve the accuracy of the deterministic prediction model.