Interval forecasting of renewable power generation

• Main Author: Luo, Lingfeng
• Other Authors: Xu Yan
• Format: Theses and Dissertations
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electric power::Production, transmission and distribution
• Online Access: http://hdl.handle.net/10356/78411
• Institution: Nanyang Technological University

With the development of renewable energy power generation industry, effective prediction of renewable energy generation is an important issue that modern power grids are facing. Solar power generation is an important part of renewable energy generation. In this project, solar incident radiation (SIR) is used as training and test data for research. By using long short term memory (LSTM) to train network parameters, the results of point forecasting are obtained which is then converted into prediction intervals by quantile regression method. Considering the uncertainty of SIR, the prediction interval is more effective than the conventional point forecasting results. Various LSTM framings are used in this project for comparison and analysis. The conclusions have a guiding role in solar power generation prediction