Application Of Type-2 Fuzzy Logic Systems In Wind Power Prediction

With the increasing of global energy demand and the sharp decline of fossil fuels,the wind power technology is becoming more and more important.One of them,accurate prediction of wind power is one of the effective methods for large-scale development and utilization of wind power.Accurate and reliable wind power prediction is very important to optimize the cost of power system operation and improve the reliability of power system.Short-term wind power prediction is of great significance to the safe and stable operation of power system,the dispatching of power network and the maintenance of wind turbines,it is one of the most important research directions in the field of new energy.At present,these wind power forecasting methods can be divided into physical method,statistical method,spatial correlation prediction method,combination forecasting method and so on.More advanced modern statistical methods,such as neural network,support vector machine,can describe the nonlinear relationship between input and output from the past time series,which have been applied successfully in the ultra short term or short term forecasting of wind power.Type-2 fuzzy logic systems as a powerful method of time series modeling,has been successfully applied to the forecasting of chaotic time series,wind speed forecasting,power load forecasting and traffic flow forecasting,and it has a very good application potential.Considerd the randomness and intermittence of wind power data and successful application of interval Type-2 fuzzy logic systems method in forecasting,the method is supposed to be one of the powerful tools for wind power forecasting.Furthermore,in order to avoid the problem of “rule explosion”,based on the type-2 fuzzy logic systems,the principal component analysis is used to reduce the input dimension.The main contents of this thesis are as follows:(1)The basic principle and algorithm of PCA and type two fuzzy sets are studied.At the same time,the composition of the two FLS and the algorithm realization of each component are studied.(2)The modeling problem of interval type-2 fuzzy logic system is studied.The method adjusts the parameters based on BP algorithm and reduces rules based on SVD-QR algorithm.In this dissertation,a multi-step predictive model of interval type-2 FLS with non-singleton type-2 fuzzification is established.And the feasibility and effectiveness of the proposed method is proved by the 20 min,40min and 60 min wind power forecasting.(3)Considering the problem of “rule explosion” of type-2 FLS and through studing PCA method,PCA and type-2 FLS are combined.This dissertation presents a forecasting method based on the combination of PCA method and interval type-2 FLS with non-singleton type-1 fuzzification and the combination of PCA method and interval type-2 FLS with non-singleton type-2 fuzzification.(4)In order to verify the effectiveness of the proposed methods,the dissertation applies different methods to short-term wind power in different regions,under the same conditions,the accuracy of the proposed method is higher than support vector machine(SVM)and another type-1 fuzzy logic method.Meanwhile,fuzzy rule explosion problems are solved effectively owing to the reduced fuzzy set by PCA transform,hence,it shows good application potential in the wind power forecasting field.