Time Series Forecasts Of CO2 Emission Variations In Madagascar Based On 1D-CNN

Harimino Andriamalala RAJAONARISOA, Adolphe Andriamanga RATIARISON

Abstract


The aim of this work was to use the 1D-CNN model to model the evolution of CO2 emission variations relative to the reference year 1990 in Madagascar. More specifically, the goal was to determine the optimal number of epochs for the 1D-CNN model that produces the best modeling performance. The datasets consisted of an annual time series of CO2 emission variations in Madagascar relative to 1990, covering the years from 1991 to 2022. For the experiment, the dataset was split into two parts: 80% (from 1991 to 2018) was used to train the model, and 20% (from 2018 to 2022) was used to test the model. The simulation was performed every 5 epochs, and the difference between the actual and predicted values was measured using the MAE (Mean Absolute Error) metric. The optimal number of epochs was determined based on the curve showing the evolution of the average MAE between the training and test data as a function of the number of epochs. After simulation, the minimum average MAE was observed at the 1965th epoch. The results of the 1D-CNN model forecasts, extending to a 10-step horizon, predict a more or less stationary trend in CO2 emissions relative to the reference year 1990 in Madagascar beyond 2023.

Keywords


1D-CNN; CO2; Epoch; Mean Absolute Error; Time series forecasts

Full Text:

PDF

References


Ritchie, H., Rosado, P., & Roser, M. (2023). CO₂ and greenhouse gas emissions. Our world in data.

Friedlingstein, P., O'sullivan, M., Jones, M. W., Andrew, R. M., Hauck, J., Olsen, A., ... & Zaehle, S. (2020). Global carbon budget 2020. Earth System Science Data Discussions, 2020, 1-3.

Le Quéré, C., Jackson, R. B., Jones, M. W., Smith, A. J., Abernethy, S., Andrew, R. M., ... & Peters, G. P. (2020). Temporary reduction in daily global CO2 emissions during the COVID-19 forced confinement. Nature climate change, 10(7), 647-653.

Watch, C. (2023). Historical GHG Emissions (1990–2020). Washington, DC: World Resources Institute.

Markova, M. (2022, April). Convolutional neural networks for forex time series forecasting. In AIP Conference Proceedings (Vol. 2459, No. 1). AIP Publishing.

Guessoum, S., Belda, S., Ferrandiz, J. M., Modiri, S., Raut, S., Dhar, S., ... & Schuh, H. (2022). The short-term prediction of length of day using 1D convolutional neural networks (1D CNN). Sensors, 22(23), 9517.

Lang, C., Steinborn, F., Steffens, O., & Lang, E. W. (2020). Applying a 1D-CNN network to electricity load forecasting. In Theory and Applications of Time Series Analysis: Selected Contributions from ITISE 2019 (pp. 205-218). Springer International Publishing.

Lang, C., Steinborn, F., Steffens, O., & Lang, E. W. (2019). Electricity Load Forecasting--An Evaluation of Simple 1D-CNN Network Structures. arXiv preprint arXiv:1911.11536.

Kiranyaz, S., Avci, O., Abdeljaber, O., Ince, T., Gabbouj, M., & Inman, D. J. (2021). 1D convolutional neural networks and applications: A survey. Mechanical systems and signal processing, 151, 107398.

Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.

Masters, D., & Luschi, C. (2018). Revisiting small batch training for deep neural networks. arXiv preprint arXiv:1804.07612.

Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning (Vol. 1, No. 2). Cambridge: MIT press.

Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.

Zhang, Z., Liu, Y., Gong, Y., & Lu, H. (2019). Forecasting air pollutant concentration levels with deep learning models. Sustainability, 11(7), 2041. https://doi.org/10.3390/su11072041

Ritchie, H., Roser, M., & Rosado, P. (2020). CO₂ and Greenhouse Gas Emissions. Our World in Data. https://ourworldindata.org/co2-and-other-greenhouse-gas-emissions

Kiranyaz, S., Ince, T., & Gabbouj, M. (2021). 1D Convolutional Neural Networks and Applications: A Survey. Mechanical Systems and Signal Processing, 151, 107398. https://doi.org/10.1016/j.ymssp.2020.107398




DOI: http://dx.doi.org/10.52155/ijpsat.v53.2.7565

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Harimino Andriamalala RAJAONARISOA

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.