The Analysis Of Wind Power Fluctuation Characteristics And The Application In The Power System

Wind is a clean energy that can mitigate greenhouse effect and is widelydistributed. Wind power generation has become the main application form of windenergy, drawn attention from the whole world, and is accelerated indevelopment.Since wind power output is random and intermittent, the power systemneed to consider the impact of wind power uncertainty on the grid in planning for thewind farm groups, that including wind power characteristics, the wind sendingcapacity that is limited,the increasing difficulty that the power grid transmit networkplanning and optimization problems and so on. So to study the characteristics oflarge-scale wind power output, to predict wind farm group output status,have greatsignificance on the Establishment of high-capacity, long-distance energy transmissionchannel to achieve optimal configuration and wind power consumptive, etc. in ourcountry.To study the impact of the wind power output characteristics on the wind farmwhen the time scale has changed. The paper multiple time scales are calculated andanalyzed in different wind power output indicators at first. Without prejudice to thedistribution of the entire wind farm, search for a relatively large time scale, whichwill help improve computational efficiency and conserve resources. In addition weneed to predict the continuous output of the wind farm to contribute to the planning ofwind farms, this paper proposes the use of probability density distribution curve todescribe continuous output predicted by the wind farm cluster probability densitydistribution of thrust reversers continued to contribute to the conditions in the windfarm when the probability density distribution of the group to predict the use ofprobability density parameter prediction studies, this paper gives the probabilitydensity function of the three Weibull, Rayleigh, t location-scale distribution of windpower output condition is described in which the Weibull distribution and probabilitydistributions of wind power output is nearest. In this paper, on the time scale when study results can be used instead ofone-hour time scale length5min sampling time, the calculation analysis shows thatthis will not affect the accuracy of the calculation and can also greatly improve thecomputational speed; proposed wind farm the group continued to contribute to theprediction method, which is planning a way to predict when the overall use of thenew wind farm (group) when a certain scale based on existing wind farms on theconvergence group, which is planning a wind farm with a some practical significance.