1 Department of Applied Mathematics and Computer Science, Technical University of Denmark2 Dynamical Systems, Department of Applied Mathematics and Computer Science, Technical University of Denmark3 Department of Electrical Engineering, Technical University of Denmark4 Center for Electric Power and Energy, Department of Electrical Engineering, Technical University of Denmark
The emphasis in this work is placed on generating space-time trajectories (also referred to as scenarios) of wind power generation. This calls for prediction of multivariate densities describing wind power generation at a number of distributed locations and for a number of successive lead times. A modelling approach taking advantage of sparsity of precision matrices is introduced for the description of the underlying space-time dependence structure. The proposed parametrization of the dependence structure accounts for such important process characteristics as non-constant conditional precisions and direction-dependent cross-correlations. Accounting for the space-time effects is shown to be crucial for generating high quality scenarios.
Wind energy; Scenarios; Trajectories; Space-time; Precision matrix; Directional CAR