Study Of Wind Power Ramping Events

Under the background of national strategies, such as the energy revolution and the ’Internet Plus’, Energy Internet can deeply combine the energy and Internet. It makes Energy Internet become a new focus of the current international academic and industrial communities. Wind energy makes an unprecedented rapid development as a kind of effective and renewable energy under the background of Global Energy Internet, especially in China. According to the actual situation of wind power integration, power system operators find that the large-scale wind power can increase or decrease in a very short time period leading to the so-called’wind power ramping event’. Wind power ramping events pose a serious threat to the security and stability of power system operations, which draws more attention to experts and academic scholars all around the world. However, the domestic and abroad research about wind power ramping events is still in a fledgling exploratory stage and the domestic research is extremely rare.Based on the aforementioned research background, this paper reviews the state of the art of domestic and abroad ramping research in detail within detection, prediction, and application areas, respectively. First, the ramping detection algorithm is studied based on the optimized swinging door algorithm. Second, deterministic and probabilistic prediction methods and another forecasting method considering the power system application are proposed. Finally, the wind power ramping event is regarded as a positive flexible ramping product and involved into the novel unit commitment dispatch model. At this point, wind power ramping events are utilized in a positive way to improve the energy internet efficiency for renewable energies.Regarding ramping events, this paper reviews the state of the art of domestic and abroad ramping research for the first time in detection, prediction, and application areas, respectively. It is demonstrated that ramping event issues will be a new research topic by using detailed literature statistics.Regarding ramping event detection algorithms, an optimized swinging door algorithm is proposed to detect wind power ramping events from wind power data. The conventional swinging door algorithm is utilized to segregate wind power data. A dynamic programming is performed to optimize the segments by merging adjacent segments with the same ramp changing direction, handling wind power bumps, and post-processing insignificant-ramps intervals. Case studies show that the optimized swinging door algorithm provides significantly better performance than the conventional swinging door algorithm and less computational time than the L1-Ramp Detect with Sliding Window algorithm.Regarding ramping event forecasting methods, this paper proposes a deterministic prediction method based on the combination of the two-dictionary atomic sparse decomposition and neural networks. Then a neural network stochastic process model is proposed based on the concept and principle of the stochastic scenario generation method. Simulation results show that the proposed model can effectively simulate the actual stochastic process of measured wind power data. Moreover, predicted ramping probabilistic distribution characteristics are also analyzed. Finally, a quasi-steady-state power flow model is established considering the amount of frequency deviation. The frequency deviation value and slip correction value are involved into the Jacobi matrix to calculate the power flow of a power system. PRESF (Post-Ramp Events Steady Frequency) and APRESF (Approximate Post-Ramp Events Steady Frequency) indices are adopted to forecast ramping events.Regarding the ramping event application, a modified unit commitment model is established considering wind power ramping products and power system flexibility based on aforementioned ramping forecasting results. Ramping events are positively used as a type of flexible ramping products and included in the unit commitment formulation considering ramping capacity limits, active power limits, and flexible ramping requirements. Numerical simulations show the effectiveness of the model considering wind power ramping products, which not only reduces the spinning reserve cost and the production cost in different extent but also does not affect the generation schedules of thermal units.