Research On The Wind Ramp Events Probabilistic Forecasting And Smoothing Method

The energy-based wind power ramp event refers to the large fluctuation in wind power caused by the change of wind speed in extreme weather in a short period of time.In recent years,because of the increasing installed capacity of wind power and the increasing scale of wind power integration,the impact of wind power changes on the security and stability of power grids is growing.Therefore,the accurate forecasting and smoothing of energy-based wind power ramp events can reduce its uncertainty.The main research contents of this paper are as follows:(1)According to the research on wind power ramp events,it is known that the wind capacity can be characterized by wind energy.Therefore,the concept of energy-based ramp events is defined by using the actual wind capacity changes in a relatively large amount in a short period of time,which is different from the definition of power-based ramp events.First of all,the forecasting methods and smoothing methods of wind power ramp events at home and abroad are deeply studied,and the basic concepts of clustering algorithms and smoothing models are explained.Secondly,the paper introduces the evaluation index of event forecasting and smoothing index,which provides a reference for the forecasting and smoothing of the energy-type ramp event in this paper.(2)According to the characteristics of extreme weather events and wind speed changes,energy-type ramp events are defined.Based on statistical theories and methods,a probabilistic forecasting method for events is proposed.Based on the short-term weather forecast values,the similarity search technique of time series is studied by using K-means clustering algorithm(EB-K)of event set.The similarity index of historical ramp events and predicted events is constructed,and a ramp event probability prediction algorithm based on amplitude and rate of change is proposed.The effectiveness of the method is verified through the analysis of numerical examples.(3)According to the forecasting algorithm and results of energy-type ramp events,in order to smoothing the energy-type ramp events,a method for assessing and suppressing the capacity of a combined fire and storage assisted ramp event is proposed.Based on the constraints of ramp climbing and storage capacity of pumped-storage power station,the objective function of ramp minimizing the expected cost of stabilizing events is established.Through the improved sequential mixed integer coding method and the improved genetic algorithm with dynamic correlation mechanism,the problem of state correlation of pumped-storage power station is solved.By solving the objective function,the storage capacity of the pumped storage power station is derived and the smoothing result of the energy-type ramp event have been obtained.The results show that the method of fire-storage combined auxiliary energy-type event smoothing proposed in this paper can save the cost of the smoothing event and verify its effectiveness.