A wind farm wind speed prediction method, system, device and storage medium

Abstract

The application discloses a kind of wind farm wind speed prediction method, system, equipment and storage medium, method includes: using improved energy valley optimization algorithm IEVO optimization in adaptive noise set empirical mode decomposition CEEMDAN White noise amplitude weight and noise adding times, wind farm historical data is decomposed by CEEMDAN;The data after decomposition is preprocessed using sample entropy SE, reduces the calculation complexity of data;Optimization is carried out to the data after preprocessing, and the random forest algorithm RF feature selection of heuristic strategy is introduced, and redundant features are eliminated;Optimized data is converted into two-dimensional image from one-dimensional time series data using gram angle and field GASF, and image features are extracted;The extracted image features are sent into time series block transformer Patch TST model for wind speed prediction, and the training algorithm GPTuner based on gradient boosting decision tree is introduced in the Patch TST model to adjust model parameters;Compared with the traditional method, the accuracy of wind speed prediction can be effectively improved, and the stable operation of wind power plant is guaranteed.

Images