| Due to the change of economic development way and the influence of global climate matters,the industry of renewable energy,especially the wind energy,has achieved further development in China.However,with the increasing penetration of wind power in power grid,the threat to power grid caused by the uncertainties and fluctuations of wind power output is also growing rapidly.Nowadays,there are lots of methods aimed at day-ahead wind power forecasting,but serious forecasting deviation still exists.The resulting wind abandonment and default not only bring huge pressure to the reserve regulation of power grid,but also seriously affect the competition level of wind power in the energy market.Based on the probability distribution model and the original forecast data,this paper drawn up the secondary schedules with and without energy storage respectively.Besides,to stabilize the prediction error of wind power,an energy storage system control strategy using moving optimization coefficients methods in the period of real time,based on the ultra-short term forecasting of wind power is proposed,which is aimed at the improvement of both the wind farm actual output quality and revenue.Firstly,based on the features of the target forecasting data extracted by using cosine similarity method,the corresponding probability distribution model of the actual power is constructed.Combined with density function and setting unit price of abandoned wind and default wind power as the weight coefficients,the paper builds mathematical models with the objective of minimizing expected cost and maximizing expected income to correct the original forecast data.The result shows that: When the system requirements are strict,the comprehensive cost of the wind farm using the secondary schedule is significantly lower than that of the original schedule,and the revenue of the wind farm is also significantly improved,but the loss of abandoned wind is increased;When the system requirements are loose,the comprehensive cost of the secondary schedule is gradually the same as that of the original schedule,and when the allowable deviation is too large,the profit corresponding to secondary schedule will be less than that of the original schedule.Then,this paper analyzes the regression characteristics of the energy storage device,and assumes its regression period in the optimization process.Within the hypothetical duration range,the energy storage space is pre-allocated according to the variance of the probability distribution of each power period,following the rules of the same amount of charging and discharging.The secondary schedule model with energy storage can be constructed in the way of presetting energy storage control range,based on the non-energy storage secondary schedule model.The results of the optimization show that the comprehensive cost and wind farm revenue situation when the wind farm adopts the secondary schedule with energy storage have been improved.At the same time,the increase in energy storage capacity,the extension of the preset return period and the relaxation of system requirements can all reduce the comprehensive cost of the wind farm,but for the benefit of the wind farm,the excessive energy storage capacity will limit its improvement space.In addition to optimizing the schedule value,it is also important to formulate an intra-day control strategy.First,in conjunction with related documents,this paper formulates a wind power quality evaluation system based on the daily root mean square error and wind farm income,and an energy storage utility evaluation standard based on the number of charge and discharge conversions and operating mileage.After that,relying on the ultra-short-term power forecast,the moving optimization method is adopted to control the state of charge of the energy storage in advance to provide a margin for the compensation of the subsequent power error,and to adjust the uncontrollable part of the energy storage in combination with the wind abandonment method.The results show that the control strategy can significantly reduce the intraday root mean square error of wind power and increase the income of the wind farm.Combined with the secondary schedule,it has a significant positive impact on the improvement of the actual output quality of the wind farm. |
