Statistical scenarios forecasting method for wind power ramp events using modified neural networks

Wind power ramp events increasingly affect the integration of wind power and cause more and more problems to the safety of power grid operation in recent years.Several forecasting techniques for wind power ramp events have been reported.In this paper,the statistical scenarios forecasting method is proposed for wind power ramp event probabilistic forecasting based on the probability generating model.Multi-objective fitness functions are established considering cumulative density functions and higher order moment autocorrelation functions with respect to the consistency of distribution and timing characteristics,respectively.Parameters of probability generating model are calculated by the iterative optimization using the modified genetic algorithm with multi-objective fitness functions.A number of statistical scenarios captured bands are generated accordingly.Eventually,ramp event probability characteristics are detected from scenarios captured bands to evaluate the ramp event forecasting method.A wind plant of Bonneville Power Administration with actual wind power data is selected for calculation and statistical analysis.It is shown that statistical results with multi-objective functions are more accurate than the results with single objective functions.Moreover,the statistical scenarios forecasting method can accurately estimate the characteristics of wind power ramp events.The results verify that the proposed method can guide the generation method of statistical scenarios and forecasting models for ramp events.

Bayesian Network Based Imprecise Probability Estimation Method for Wind Power Ramp Events

Although wind power ramp events(WPREs)are relatively scarce,they can inevitably deteriorate the stability of power system operation and bring risks to the trading of electricity market.In this paper,an imprecise conditional probability estimation method for WPREs is proposed based on the Bayesian network(BN)theory.The method uses the maximum weight spanning tree(MWST)and greedy search(GS)to build a BN that has the highest fitting degree with the observed data.Meanwhile,an extended imprecise Dirichlet model(IDM)is developed to estimate the parameters of the BN,which quantificationally reflect the ambiguous dependencies among the random ramp event and various meteorological variables.The BN is then applied to predict the interval probability of each possible ramp state under the given meteorological conditions,which is expected to cover the target probability at a specified confidence level.The proposed method can quantify the uncertainty of the probabilistic ramp event estimation.Meanwhile,by using the extracted dependencies and Bayesian rules,the method can simplify the conditional probability estimation and perform reliable prediction even with scarce samples.Test results on a real wind farm with three-year operation data illustrate the effectiveness of the proposed method.