Study On The Strategy Of Planning Of The Wind Farm Output

With the large wind power connecting to power grid, volatility of wind power,instability and uncertainty in forecasting makes wind power has become the maindemand of ancillary services. Now, many countries have wind power forecastingtechniques as wind farms hair wind power technical requirements. Less thanqualified for the prediction accuracy of the wind farm will not be connected topower grid. If forced grid, wind farm must accept the fines brought by predictionbias to pay for costs related to ancillary services. So the reporting output of windfarms will directly affect their own economic gain. Therefore, the wind powerplanning report strategy has practical significance in theory and engnieeringanalysis.Based on the research and analysis of wind power prediction error probabilitycauses and distribution distribution characteristics, this paper put forward a kindof beta distribution parameter optimization modele to fit the frequency distributionof the prediction error statistics and get the wind power prediction error probabilitydistribution function and the fluctuation interval estimation fitting, at the same timeusing beta distribution, normal distribution model analysis comparing theeffectiveness of the model. And on this basis, this paper, drawing on recent windpower capacity of wind power implementation of reporting system and method forassessment and reporting accuracy domestic remedies for ancillary services, Windpower business as a target for maximum revenue expectations of independenteconomic body, get the best strategy of planning of the wind farm output.Finally, taking it as an example that the actual power of a wind farm in InnerMongolia and forecast data based on the power neural network prediction method,the simulation verified the effectiveness of the model under different intervals anddifferent predictive power penalty coefficient.