Power forecasting of photovoltaic system using environment data and weather information

Solar energy is a very important technology for the development of renewable energy in the world. However, solar energy is with the disadvantage of intermittent power generation and cannot be supplied around all day. In order to increasing the accuracy of photovoltaic modules in forecasting the power generation, this study is applied the Internet of Things technology to detect the environmental data of photovoltaic power generation, and uploads the detected data to the database via the internet. These are applied to train and update the weights of fuzzy neural networks for modeling and forecasting. This study attempts to compare the impact of only using environmental information and adding weather forecast information on the training results.