Wind power prediction and improvement algorithm based on multi-objective collaborative training

Abstract

Under the ′dual carbon′ goal, the transformation of the power system is accelerating, and wind power prediction technology is of great significance to the construction of a new power system with a high proportion of new energy. In order to improve the accuracy and robustness of wind power prediction, an numerical weather predicition (NWP) implicit correction algorithm based on multi-objective collaborative training is proposed. Firstly, the necessity of NWP correction and the problems of the two-step prediction method based on NWP explicit correction are analyzed. Then, aiming at the problems existing in the two-step prediction method, the optimization method based on multi-objective collaborative training uses the neural network to perform NWP implicit correction, train the model in an end-to-end manner, and realize the functions of NWP implicit correction and wind power prediction at the same time. Combined with the measured data of a wind farm, the specific calculation case analysis proves that the proposed algorithm has an improving effect on short-term, medium- and long-term wind power prediction. In addition, the algorithm only requires one network and avoids secondary calculation, saving computing and storage costs.

Keywords

• wind power forecasting
• numerical weather predicition (nwp) implicit correction
• neural networks
• improvement algorithm
• multi-objective collaborative training
• two-step forecasting method