RT info:eu-repo/semantics/masterThesis
T1 A Multivariate Prediction Model for Short-Term Photovoltaic Plant Generation Using Bi-LSTM and CNN
A1 Ivanov Kurtev, Kiril
A2 Máster Universitario en Energías Renovables
AB The short-term prediction of the energy produced by a photovoltaic plant is a widelystudied topic, and it is an important issue for the stability of the grid and its correct operation,as well as for reducing the operating costs and increasing the lifetime of the elements thatmake up it. The creation of a tool to more accurately predict the solar generation of thePV plant, specifically the prediction of ramps 5-10 minutes in advance. In this work, amultivariate prediction model is presented that combines images, historical production data,and solar position at each moment. The model consists of two parts: image processing, witha convolutional neural network (CNN) and time series processing using a Bidirectional LongShort-Term Memory (Bi-LSTM) capable of detecting long-term nonlinear features. CNNswill be trained to automatically detect the relationship between the images taken of the skyand cloud movement and the current power of the solar array. Then, the recurrent neuralnetworks (RNNs) created will be used to give rise to a 5-minute prediction and a 10-minuteprediction. The prediction results are compared using different error metrics, like skill scoreand the mean squared error (RMSE).
YR 2022
FD 2022
LK http://riull.ull.es/xmlui/handle/915/31743
UL http://riull.ull.es/xmlui/handle/915/31743
LA en
DS Repositorio institucional de la Universidad de La Laguna
RD 08-may-2024