Performance assessment of SARIMA, MLP and LSTM models for short-term solar irradiance prediction under different climates in Morocco

Photovoltaic (PV) production is highly dependent on global solar irradiance (GHI) and often experiences irregular fluctuations. With the continuous increase in PV penetration rates, GHI forecasting methods are becoming important to ensure optimal management of the energy produced. In this study, thr...

Full description

Saved in:

Bibliographic Details
Published in:	International journal of ambient energy Vol. 44; no. 1; pp. 334 - 350
Main Authors:	El Alani, Omaima, Hajjaj, Charaf, Ghennioui, Hicham, Ghennioui, Abdellatif, Blanc, Philippe, Saint-Drenan, Yves-Marie, El Monady, Mohammed
Format:	Journal Article
Language:	English
Published:	Taylor & Francis 31-12-2023
Subjects:	domain_math.appl Engineering Sciences Environmental Sciences Forecasting machine learning Mathematics Ocean, Atmosphere PV production Sciences of the Universe solar irradiance
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Photovoltaic (PV) production is highly dependent on global solar irradiance (GHI) and often experiences irregular fluctuations. With the continuous increase in PV penetration rates, GHI forecasting methods are becoming important to ensure optimal management of the energy produced. In this study, three forecasting models - Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), and Seasonal Autoregressive Integrated Moving Average (SARIMA) were evaluated to predict GHI one day ahead. Ground measurement data from four sites in Morocco and weather forecasts from the Global Forecast System (GFS) were used to perform the forecast. Results show that forecasts based on MLP and LSTM are more accurate than SARIMA and persistence even if under complicated weather conditions. For clear days with low variability, the RMSE for LSTM, MLP M1, and MLP M2 are 18.58, 12.97, and 45.20 W/m 2 . For cloudy days with high variability, the RMSE are 103, 60.80 W/m 2 , and 89.17 W/m 2 , respectively.
ISSN:	0143-0750 2162-8246
DOI:	10.1080/01430750.2022.2127889