Research On Applicaton Of Power Forecasting And Hybrid Energy Storage In Wind Generation System

Demands for new energy are increasingly growing because of energy depletion and environmental pollution. Compared with other new energy, wind power is easier to realize its industrial development as a result of its high technology maturity and low cost. However, development of wind power is confined by the influence of its fluctuation and randomness. There may have two methods to improve the steadiness of wind power integration. One is accurate prediction of wind power. System reserve capacity can be reduced through adjusting the scheduling plan on basis of prediction result. The other is the application of energy storage. Power quality and wind curtailment problem will be improved with the help of storage devices. This paper addressed to wind power prediction and the coordination control of hybrid energy storage, main contents are as follows.Firstly, in order to reduce system reserve capacity based on accurate prediction result, three kinds of prediction algorithms are introduced. Supper short-term forecast is implemented by time series analysis method. Then back propagation neural networks is used owing to nonlinear data processing ability and high robustness. At last, particle swarm optimization algorithm is used to adjust the network weights, taking advantage of its balance ability between the global and local to avoid trapping into local minima. The error analysis result proves the effectiveness and accuracy of PSO-based BP algorithm.Secondly, to assure system steadiness under the condition of wind power fluctuation, direct drive permanent magnet synchronous generator model is built, and the battery-ultracapacitor hybrid energy storage is confirmed to replace single storage element. Moreover, model of energy storage system is built, consisting of battery, ultracapacitor, bi-directional DC-DC convertor and DC-AC invertor, which makes it possible to research on control strategy of hybrid energy storage system.Thirdly, for purpose of stabilizing power fluctuation, it presents a control strategy for hybrid energy storage system based on low-pass first order filter after researches on the characteristics of storage elements. After that, PQ control is used by DC-AC invertor to realize power compensation of battery and constant power control strategy is applied by DC-DC convertor to realize power compensation of ultracapacitor. At last, with the help of Matlab/Simulink, the simulation result demonstrates the effectiveness of the control method for reduction of the power fluctuation.