Solar irradiance prediction using weather forecasts by Long Short Term Memory(LSTM)

Vijayan, Viniyta (2021) Solar irradiance prediction using weather forecasts by Long Short Term Memory(LSTM). Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Solar irradiance prediction using weather forecasts by Long Short Term Memory(LSTM).pdf - Submitted Version Download (7MB)

Abstract

Solar irradiance prediction is for maximizing energy saving costs and providing high power quality in electrical power with distributed solar photovoltaic generations. The irradiance data can not be obtained because of expensive solar irradiance meter and the irradiance forecasting data are often unavailable. There are many methods is introduced that we can use to predict solar irradiance for the hourly day ahead by using weather forecasting data. This project studies a technique using Long Short Term Memory (LSTM) to predict solar irradiance using Deep learning toolbox network. The weather parameter from Faculty of Electronic and Computer Engineering (FKEKK), Malacca. Three algorithms namely Levenberg-Marquardt (LM), Bayesian Regularization (BR) and Scaled Conjugate Gradient (SCG) are used in the weather forecasting model. Their results are compared based on their performance of Mean Square Error (MSE) , Root Mean Square Error (RMSE) and Regression (R). Therefore, by giving predicted weather parameter as input, the model will give solar irradiance prediction.

Actions (login required)

View Item