Distributed clean energy electric quantity prediction method under dual-carbon target

The invention discloses a distributed clean energy electric quantity prediction method under a dual-carbon target, aims at solving the problem of how to predict the distributed clean energy electric quantity, and belongs to the technical field of clean energy. The method comprises the following steps: S1, acquiring historical data of distributed clean energy in past 12 hours, wherein the historical data comprises operating parameters and photovoltaic power of the distributed clean energy; s2, extracting features of the historical data, performing random forest feature screening on the extracted features, and constructing a training set and a test set; and S3, establishing an LSTM multi-step prediction model, training the LSTM multi-step prediction model by using the training set, testing the LSTM multi-step prediction model by using the test set after the training is completed, if the accuracy rate reaches the standard, outputting the LSTM multi-step prediction model for predicting the photovoltaic power in t