BP neural network photovoltaic power prediction method based on information fusion theory

• Main Authors: ZHAO XINXIN, YANG HAIYAN, HU-RONG CHAOHUI, HUANG ZHI, YANG MINGSHENG, XIA XIANGYANG
• Format: Patent
• Language: Chinese; English
• Published: 03.08.2018
• 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; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a BP neural network photovoltaic power prediction method based on an information fusion theory. As a new energy source having the advantages of environmental protection, sustainability and short construction period, photovoltaic power generation has become a major force in power generation. However, due to the intermittence, randomness and volatility of photovoltaic power generation, a high ratio of effective access is difficult to guarantee, thus a certain influence on safe operation and dispatching of a power grid is caused. Precise photovoltaic power prediction can effectively solve the problem and accelerate the development of photovoltaic power generation. The BP neural network photovoltaic power prediction method fully considers influence factors of photovoltaic power generation and fuses the influence factors into an influence factor lambda, utilizes the advantage of adjustable structure of a BP neural network to predict moments with great fluctuations precisely, can realize eff