International Journal of Automation and Safety
Volume 2, Numéro 2, Pages 20-23
2024-12-31
Predicting Hdd And Cdd In Batna, Algeria: A Comparative Study Of Lstm And Prophet Models
Authors : Bouzid Nassim . Aksa Karima . Mouss Mohammed Djamel .
Abstract
In this study, the Heating Degree Days (HDD) and Cooling Degree Days (CDD) of the Wilaya of Batna are estimated by comparing two prediction models: Facebook Prophet (Prophet) and Long Short-Term Memory (LSTM). These indicators are crucial for improving energy efficiency by optimizing the management of heating and cooling systems. The analysis is based on daily climate data over a 20-year period, from January 2000 to December 2020. Precision metrics including the coefficient of determination (R²), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) were used to assess the models' performance. The findings show that, in comparison to the Prophet model, the LSTM model predicts HDD and CDD values more accurately. The performance of the LSTM model is as follows: HDD RMSE of 1.8663, MAE of 1.4042, and R² of 0.8896; CDD RMSE of 1.0413, MAE of 0.6235, and R² of 0.8904. In comparison, with an HDD RMSE of 2.4297, MAE of 1.7742, and R² of 0.8127; a CDD RMSE of 1.4955, MAE of 0.8296, and R² of 0.7737, the Prophet model produces less accurate findings. This study highlights that the LSTM model is better suited for predicting HDD and CDD values in Batna, which could enhance energy planning and optimize the efficiency of heating and cooling systems. Consequently, using the LSTM model may contribute to reducing energy consumption.
Keywords
Cooling Degree Days (CDD) ; Energy Efficiency ; Facebook Prophet (Prophet) ; Heating Degree Days (HDD) ; Long Short-Term Memory (LSTM)
Les articles similaires
Khledj Meryem
.
Mansouri Hadjmoussa
.
Hela Borgi
.
pages 293-307.
Lagoune Salim
.
pages 13-27.
عنصري محمد
.
عقبان زينب
.
ص 481-499.
Bekhti Sara
.
Benamor Silia
.
Sellami Asma
.
pages 15-25.
