The Study On Wind Power Forecasting And Wind Farm Storage Capacity Optimization

As the depletion of energy and pollution of environment are becoming more and more serious, nowadays, people are putting more emphasis on the research and utilization of the renewable and clean energy. Wind power is one of the renewable and clean energy and that two problem will be alleviated if the wind power is utilized. It is necessary to take a deep research because the wind power shortage is emerging in the whole process.This paper had focus on two aspects, the first one is the prediction of wind power, the second is the optimization of energy storage of the wind power station.This paper introduced the development of wind power from three different range, the global, domestic, Shandong Province. We showed the current status of wind power development through a large number of statistical data.For wind power prediction, this paper first carries on the analysis to the historical data, from the perspective of statistics analysis of their probability distribution. Then using DFITTOOL in MATLAB to fitting the distribution. We found the t-location scale is good because it has long tail. On the basis, wind power prediction methods was studied. We studied the ARIMA module based on time series and the BP neural network module based on artificial intelligence. Then analysis the data from wind farm and the optimum parameters of the models were determined using SPSS statistical software. And the modules was tested with part of data. Finally, the comparison of accuracy, the qualified rate of the multiple angles of the two kinds of model.According to the wind farm energy optimization problem, a system control model named first-order low-pass filter was put forward for wind farm storage, which was based on the analysis of the storage battery output model. At the same time put forward the system capacity allocation method which can make the maximum economic benefits, consider the power benefits,environmental benefits and so on.Finally, the work of this thesis was summarized, and the future work was prospected.