Wind power short-term power prediction method based on deep learning and multiple integration

• Main Authors: GAO DELAN, SHI RUXIN, CAO QINGCAI, WANG JUAN, TADAO, ZHANG JIANXIN, LIU XIANRONG, TANG HONGFEN, CAO SHANQIAO, ZHANG SHUXIANG, WU LIDONG, XUN JIAMENG, ZHANG SHUXIAO, XU ZHIXUAN, GUO XUFENG, ZHANG LIXING
• Format: Patent
• Language: Chinese; English
• Published: 17.05.2022
• Subjects: CALCULATING; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTINGELECTRIC POWER; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR; ELECTRICITY; GENERATION; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS; PHYSICS; PRINTED CIRCUITS; SYSTEMS FOR STORING ELECTRIC ENERGY

The invention discloses a wind power short-term power prediction method based on deep learning and multiple integration. The wind power short-term power prediction method comprises the following steps: step 1, acquiring meteorological data of an area where wind power to be predicted is located by using a data collection device; step 2, the data collection device preprocesses the acquired meteorological data through a preprocessing module; 3, aiming at different variables of the meteorological data, carrying out normalization processing by adopting a standard method; step 3, analyzing correlation among different variables of the meteorological data; 4, designing a deep and multiple integrated neural network structure; 5, training the deep and multiple integrated neural network by using the training sample data; and 6, in a wind power short-term power prediction module, predicting the wind power short-term power of the test sample by using the trained depth and multiple integrated neural network model. Accordin