Research On The Application Of Fuzzy Time Series Prediction Model In Wind Power Generation In China

With the advent of the new energy era,the low-carbon energy development model is accelerating the transformation,and the green and diverse energy supply system is accelerating,Wind power is a clean,low-carbon,safe and efficient new energy,it has made rapid development achievements in the world.The accurate prediction of wind power is the basis and key of the effective utilization of wind energy,which can effectively avoid the occurrence of "abandon wind" phenomenon,avoid the waste of wind resources,and help to improve the regulation ability and operation efficiency of the national electricity system,and also play a certain role in promoting the construction and development of the national.In this paper,we select the FTS prediction model and the IFTS forecast model to forecast the wind power in China,which has a series of notable advantages: low demand for sample data,no need to meet many restrictions,can predict the data of non-genera value samples and the development trend is not clear sample data.In view of this,the main contents of this article are as follows:Firstly,the fuzzy time series prediction model(FTS)is introduced and optimized,this paper mainly improves the division of domain interval,first divides the domain by FCM clustering algorithm,and then divides the interval by two times according to the number of historical data in the interval.Remove intervals that do not contain any historical data.The classification method makes the historical data subordinate to the corresponding interval higher,so that the heuristic fuzzy relation group of each historical data corresponding to the fuzzy set is not empty.Through the prediction of the enrollment of Alabama University,the improved method of domain partitioning makes the prediction precision of the model improved significantly,thus verifying the feasibility of this method.Finally,we use the improved FTS model to forecast the wind power in our country in the short term.Secondly,a first-order and high-order intuitionistic fuzzy time series prediction model(IFTS)is constructed.In the division of domain,the improved two-time domain partitioning method is adopted;In the process of intuitionistic fuzziness of historical data,Gaussian function is used to determine membership degree and nonmembership degree;In the establishment of the prediction rule,the Mamdani implication operator is used to make intuitionistic fuzzy multiple inference and intuitionistic fuzzy multidimensional inference.Using heuristic multiple reasoning principle and multidimensional reasoning principle to synthesize the inference,then the weighted average value method is used to get the prediction value,and the low order and high order IFTS forecasting model are used to forecast the wind power generation in our country respectively.Finally,in order to verify the effectiveness of FTS prediction model and IFTS prediction model to the wind power generation prediction in our country,the paper constructs theclassical time series decomposition forecast model.The results show that the FTS model and the IFTS model have better prediction effect,and the high order IFTS prediction model has the best prediction performance.These two predictive models can be more objective and comprehensive in the reaction time series of deterministic and uncertain factors,more adaptable.Therefore,it is effective to use FTS prediction model and IFTS forecast model to forecast the wind power in China.