Study On The Modeling Method Of New Energy Power Time Series Based On Fluctuation Characteristics

With rapid increasing of new energy power generating capacity, the influence of new energy power on power system planning and operation become more and more severe. Time series modeling of new energy power is the basis of evaluating the influence accurately of new energy power on power system, which is very important to annual new energy dispatch planning, new energy installed capacity optimization.On the basis of investigating wind power/photovoltaic(PV) power time series fluctuation charcteristics, time series modeling and new energy acceptance capacity evaluation method, this paper proposes a wind power time series modeling method considering fluctuation characteristics, a PV power time series modeling method based on clear sky model and medium and long term sequential production simulation considering new energy power stochastic characteristics, provides the technical support for large-scale new energy dispatch and operation. The main contribution of this paper are as follows:First, A new method of fluctuation process is proposed to describe the stochastic and varition characteristics of wind power. Wind power fluctuation process characteristics is analyzed, wind power fluctuations profiles can be fitted by Gauss function perfectly and it can quantitative express fluctuation variation laws. Also wind power fluctuations categories can be identified by visualization Self-Organizing Map (SOM) clustering algorithm. Finally analyses the time distribution characteristics of all categories of fluctuations.Second, based on wind power fluctuation characteristics, a new modeling method of wind power time series is proposed. By using statistical method and sequential sampling method, wind power time series can be simulated. The validity of simulated time series is tested and results compared with other methods show that this method is progressive.Third, a new PV power time series modeling method is proposed, PV power decomposition model which includes certain output based on clear sky model and uncertain output based on weather characteristics is put forward and PV power characteristics under various weather process are analyzed. On above basis, PV power time series is simulated by statistical parameters and sampling method, through testing the characteristics of simulation results, the validity of this method is verified.Finally annual time sequential production simulation model is built by taking one provincial power grid as example and by using Monte Carlo method to carry on production simulation repeatedly, the expect of new energy limited power are acquired.Besides a sensitivity analysis is also carried out based on the indices of reserve capacity,load and wind source.