Integrated learning photovoltaic power generation prediction method for building integrated energy management

• Main Authors: XU HAOZE, CHEN FUDONG, ZHU XIAOMING, WU YAOYANG
• Format: Patent
• Language: Chinese; English
• Published: 18.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; 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 discloses an integrated learning photovoltaic power generation prediction method for building integrated energy management, which is a Stacking integrated learning method based on an XGBoost element learner, and comprises the following specific implementation steps: obtaining various characteristic variables of building photovoltaic power generation as a data set of a model, preprocessing the data set, and dividing the data set; lSTM and LSSVM are adopted as base learners of ensemble learning, a K-fold cross-check method is used, model training is carried out, a prediction result is obtained, meanwhile, prediction is carried out on a test set, and after prediction values are averaged, a new test set is obtained; the K-Fold verification set prediction results of the LSTM and the LSSVM are used as a training set of a meta-learner XGBoost, model training is carried out again, and a final prediction result is obtained on a new test set; and the prediction accuracy is evaluated through RMSE, and the