Research On Wind Power Prediction Considering Dynamic Temporal Spatial Relationship Of Wind Speed

Wind power is intermittent and fluctuant, large scale grid integration of wind power brings a severe challenge to the safe and stable operation of electric power system. As the best way to solve the problem caused by grid integration of wind power, prediction of wind output power can provide important reference and support for the control and operation of the power grid, effectively increase the wind power quasi access capacity and improve the security and economy of the system operation.So, it's of great significance to study the method to forecast wind output power.At present, the research on the method to forecast wind power at home and abroad is mainly focused on the prediction method based on statistics,these method need a lot of historical data support, the prediction error increases with the increase of the time scale, and the prediction is poor when weather mutation,so this method is used for ultra-short term forecasting of wind power.But, there is little research on the physical prediction method to forecast wind output power.Due to the above situation, this paper proposes a wind power forecasting method, which considers the dynamic temporal and spatial relationship of wind speed,taking NWP data as input, this method forecasts wind output power by correcting NWP data,calculating the wind speed distribution, and converting wind speed and power:First, we analysis the wind farm local effect and the influence on low field caused by wind direction change,and establish the local effect model of wind farm and the terrain identification model considering wind direction. Through the model, we analyze the physical process of air flow,and achieve the goal of realizing the correction of NWP data,and improving the prediction accuracy.Then,we analyze the distribution characteristics of wind speed in the wind turbine based on three aspects: space, time and dynamic change, and propose a wind speed model,which considers the dynamic spatial and temporal distribution characteristics of wind speed in the wind farm.The model can calculate the spatial and temporal distribution of the wind speed in the forecast period through simulating effect of wind turbine and influence caused by wind direction changing, and forecast wind speed of each wind turbine in the wind farm.Finally, we study and compare three methods of modeling wind speed power curve,and select the wind speed power curve drawn by Bean.We can forecast the output power of each wind turbine in the wind farm by combining the curve and the predicted wind speed.Programing to achieve the above process in the MATLAB platform, and forecast the wind power of a wind farm in the next twenty-four hours. The numerical examples show that the proposed method has the advantages of fast calculation speed, high engineering practicability, strong applicability for all kinds of weather conditions, and can be used to realize the short-term forecasting of wind power output.