Short-term power prediction of wind farm cluster considering spatio-temporal correlations

Wind energy is an important low-carbon energy source, which has many advantages such as low cost, clean and renewable. The precise prediction of wind power holds immense importance in facilitating the grid-connected operational efficacy of large-scale wind farms. Conventional methods for wind power forecasting fail to address the temporal and spatial interdependence between the output of different wind turbines in a wind farm, which leads to the low accuracy of the forecasting model. This paper proposes to establish power forecasting models considering time-space correlation, to analyze the correlation of wind speed data in space between adjacent wind farms by using correlation coefficient method, and then to perform short-term wind power forecasting based on Back propagation (BP) neural network. Finally, the actual operation data from adjacent wind farms in a specific region of Xinjiang is utilized to demonstrate and validate the feasibility and accuracy of the model.