New energy power prediction feature recursive screening method based on random forest

• Main Authors: LI ZHUOHUAN, MA XIYUAN, LI PENG, XI WEI, CHEN YANSEN, ZHOU CHANGCHENG, PAN SHIXIAN, CHENG KAI, BAO TAO, HU XUDONG
• Format: Patent
• Language: Chinese; English
• Published: 27.06.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention relates to a new energy power prediction feature recursive screening method based on a random forest. The method comprises the following steps: acquiring initial meteorological characteristics of a new energy power system; according to the correlation between the initial meteorological characteristics and the generated power of the new energy power system, screening out a first meteorological characteristic from the initial meteorological characteristics; determining a second meteorological feature from the first meteorological features according to the importance of the first meteorological features on power prediction of the new energy power system; respectively taking the first meteorological feature and the second meteorological feature as training samples, training a to-be-trained power prediction model, and determining a target meteorological feature from the first meteorological feature and the second meteorological feature according to the prediction precision of the obtained pre-trained