Optimal Dispatching Of Power System Considering Hidden Prediction Error Of Wind Power

With the continuous increase of wind power penetration rate in the power system,the traditional power system is gradually evolving to the new power system with high proportion of wind power.However,the randomness and volatility of wind power bring great challenges to the dispatching and operation of power system.Therefore,it is of great significance for improving the security and stability of the system to accurately reveal the real-time fluctuation characteristics of wind power,improve wind power prediction accuracy and formulate power system dispatching plans which can cope with the randomness and volatility of wind power.To evaluate the instantaneous prediction error of wind power more accurately,hidden prediction error of wind power and corresponding evaluation indexes are established.The established hidden prediction error can make up for the performance disadvantages that the existing wind power prediction error index system cannot accurately evaluate the real-time characteristics of prediction error so that the prediction error index system is improved.Finally,case studies using actual wind farm power data verify the proposed prediction error and evaluation indexes can more comprehensively and accurately evaluate the prediction error.The improvement of evaluation effect is more obvious especially when the wind power has the characteristics of extreme fluctuations within the prediction time step.To reduce the hidden prediction error of ultra-short-term wind power prediction and overcome the difficulty that the prediction model with fixed time-resolution cannot accurately predict the extreme wind power fluctuations,an ultra-short-term wind power prediction model with adaptive time-resolution is proposed.In this model,the adjustment time of time-resolution is determined by evaluating the fluctuation magnitude of hidden prediction error in the prediction target time interval and the adjustment rules are formulated by mining the regularities of fluctuation rate of historical wind power data and establishing the interval grouping optimization model.Then,by coupling the adjustment time and rules into the prediction model of back propagation neural network,the rolling prediction with adaptive adjustment of time-resolution is achieved.Finally,extensive tests using actual wind farm power data demonstrate that the proposed prediction model can adaptively adjust the time-resolution to accurately track the real-time fluctuation of wind power,which greatly reduces the prediction error.Especially when the fluctuation of wind power is extremely violent,the prediction error is reduced more significantly and its maximum hidden error rate can be reduced by more than 40%.To make the wind power uncertainty more accurately described by the power system dispatching model and make the intra-day dispatching plans more effectively cope with the uncertainty caused by the instantaneous fluctuation of wind power,a robust optimization dispatch model considering wind power hidden prediction error is proposed.In this model,the wind power prediction values with adaptive time-resolution are processed by using a numerical conversion method of linear interpolation first and averaging second to obtain the wind power baseline scenario.Then a wind power deviation scenario set construction method based on hidden prediction error is established to form the deviation scenario set.By combining the baseline scenario with the deviation scenario set,the wind power scenario beam with high accuracy and effectiveness is constructed and applied to the robust optimization dispatch model to formulate intra-day dispatching plans which take into account wind power hidden prediction error.Finally,IEEE-30 node system and actual wind farm power data are adopted to conduct case studies.The comparison results demonstrate that the wind power scenario beam constructed by the proposed dispatch model can more accurately and effectively describe the uncertainty of wind power,and that the model can formulate more efficient dispatching plans under different wind power fluctuation conditions.