Research On The Application Of Ultra-short-term Wind Power Forecasting And Energy Storage System In Wind Farm

With the increasing consumption of fossil energy,wind energy has been increasingly valued as a renewable energy source,and the proportion of wind power in the power grid has continued to increase.However,due to the random,intermittent,and fluctuating characteristics of wind power,it will bring great challenges to the power quality,and safety and stable operation of the power grid.Wind power forecasting is one of the critical technical methods to solve the above problems.Due to various reasons,the prediction accuracy especially the day-ahead power forecast of some wind farms in China cannot meet the requirements.Using the energy storage system(ESS)to track wind power schedule output is an effective way to reduce the prediction error indirectly.In this paper,ultra-short-term forecasting of wind power forecasting was studied,and ESS was used to track the day ahead wind power schedule output.The actual data of a wind farm through field investigation was obtained,it’s used to analyze the wind power output characteristics.Aiming at the problem of ultra-short-term wind speed prediction,a Kalman modified multi-step wind speed prediction model was proposed.The model was used to correct the numerical weather prediction(NWP)wind speed error by combining real data.The ARIMA model was used to predict wind speed with historical wind speed data as a comparison,and actual data from a wind farm was used to verify the models.The multi-step wind speed corrected by Kalman filter,ARIMA predicted wind speed,and NWP wind speed were used to predict wind power modeled by support vector regression(SVR).The results showed that the Kalman filter multi-step corrected-SVR model can improve wind power prediction accuracy effectively.ESS was used to improve the ability of day-ahead wind power schedule output tracking based on the relationship between the ESS’s state of charge(SOC),ultra-short-term predicted power and the day ahead wind power schedule output,to formulate the charging and discharging strategy of the ESS.The objective function weighted by the planned output tracking,smoothing power fluctuation of the combined system and the charging and discharging capacity of the ESS was proposed.Through point-by-point rolling optimization,the planned output tracking ability of the ESS was improved,and an example was used to verify the feasibility of the strategy.