A Pipeline for Multivariate Time Series Forecasting of Gas Consumption in Pelletization Process
DOI:
https://doi.org/10.19153/cleiej.28.3.2Keywords:
Machine Learning, AutoML, Neural Networks, Feature Selection, Evolutionary AlgorithmAbstract
Gas consumption is a critical aspect of the pelletizing process, directly influencing operational costs and environmental impact. This study investigates the application of a multivariate time series forecasting pipeline for predicting gas consumption in pelletizing plants. The pipeline comprises: (i) data preprocessing, (ii) converting the dataset into a tabular format using a sliding window technique, (iii) applying feature selection methods, and (iv) employing machine learning tuned via AutoML. The methodology was tested on a dataset with 45 operational parameters collected over 90 days from an industrial plant, with predictions evaluated using Root Mean Squared Error (RMSE). In step (iii), twelve features were identified as the most relevant based on the Random Forest importance index. In the final stage, two AutoML approaches were employed: neural architecture search using AutoKeras and the DEAP (Distributed Evolutionary Algorithm) framework. The neural network architectures tested included MLP, RNN, LSTM, and Conv1D. The best performance was achieved by the DEAP framework combined with LSTM networks, which yielded an RMSE of 0.33. Although AutoML did not outperform the statistical model in terms of RMSE values, regarding training time, AutoML models were significantly more efficient than the statistical approach, optimizing computational resource usage and enabling faster model adjustments. These findings confirm the generalization capability of the pipeline, demonstrating its applicability across different industrial environments.
References
M. Mudelsee, “Trend analysis of climate time series: A review of methods,” Earth-science reviews, vol. 190, pp. 310–322, 2019. [Online]. Available: https://doi.org/10.1016/j.earscirev.2018.12.005
S.-Y. Shih, F.-K. Sun, and H.-y. Lee, “Temporal pattern attention for multivariate time series forecasting,” Machine Learning, vol. 108, no. 8, pp. 1421–1441, 2019. [Online]. Available: https://doi.org/10.1007/s10994-019-05815-0
A. Gonzalez-Vidal, F. Jimenez, and A. F. Gomez-Skarmeta, “A methodology for energy multivariate time series forecasting in smart buildings based on feature selection,” Energy and Buildings, vol. 196, pp. 71–82, 2019. [Online]. Available: https://doi.org/10.1016/j.enbuild.2019.05.021
X. Yuan, J. Yuan, T. Jiang, and Q. U. Ain, “Integrated long-term stock selection models based on feature selection and machine learning algorithms for China stock market,” IEEE Access, vol. 8, pp. 22 672–22 685, 2020. [Online]. Available: https://doi.org/10.1109/ACCESS.2020.2969293
R. Zebari, A. Abdulazeez, D. Zeebaree, D. Zebari, and J. Saeed, “A comprehensive review of dimensionality reduction techniques for feature selection and feature extraction,” Journal of Applied Science and Technology Trends, vol. 1, no. 1, pp. 56–70, 2020. [Online]. Available: https://doi.org/10.38094/jastt1224
V. M. de Oliveira, K. S. Komati, and J. O. Andrade, “Implementing neuroevolution for gás consumption forecasting in the steel industry,” in 2024 L Latin American Computer Conference (CLEI), 2024, pp. 1–10. [Online]. Available: https://doi.org/10.1109/CLEI64178.2024.10700389
A. F. F. d. Cruz, E. B. Haagensen, L. R. Zanon, W. W. D. Regattieri, and M. M. Machado, “Aumento da eficiência energética em fornos de politização,” in Anais do 6° Simpósio Brasileiro de Aglomeração de Min´erios, 2018, pp. 127–139. [Online]. Available: https://doi.org/10.5151/2594-357X-31608
E. M. d. Santos, M. T. W. Fag´a, C. B. Barufi, and P. L. Poulallion, “Gás natural: a construção de uma nova civilização,” Estudos Avançados, vol. 21, pp. 67–90, 2007. [Online]. Available: https://doi.org/10.1590/S0103-40142007000100007
V. Pavithra and V. Jayalakshmi, “Hybrid feature selection technique for prediction of cardiovascular diseases,” Materials Today: Proceedings, vol. 81, pp. 336–340, 2023, international Virtual Conference on Sustainable Materials (IVCSM-2k20). [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2214785321022549
A. Garcia-Hiernaux, J. Casals, and M. Jerez, “Fast estimation methods for time-series models in state–space form,” Journal of Statistical Computation and Simulation, vol. 79, no. 2, pp. 121–134, 2009. [Online]. Available: https://doi.org/10.1080/00949650701617249
R. Wan, S. Mei, J. Wang, M. Liu, and F. Yang, “Multivariate temporal convolutional network: A deep neural networks approach for multivariate time series forecasting,” Electronics, vol. 8, no. 8, p. 876, 2019. [Online]. Available: https://doi.org/10.3390/electronics8080876
D. Baymurzina, E. Golikov, and M. Burtsev, “A review of neural architecture search,” Neurocomputing, vol. 474, pp. 82–93, 2022. [Online]. Available: https://doi.org/10.1016/j.neucom.2021.12.014
R. Miikkulainen, J. Liang, E. Meyerson, A. Rawal, D. Fink, O. Francon, B. Raju, H. Shahrzad, A. Navruzyan, N. Duffy et al., “Evolving deep neural networks,” in Artificial Intelligence in the Age of Neural Networks and Brain Computing. Elsevier, 2019, pp. 293–312. [Online]. Available: https://doi.org/10.1016/B978-0-323-96104-2.00002-6
H. Jin, Q. Song, and X. Hu, “Auto-keras: An efficient neural architecture search system,” in Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, 2019, pp. 1946–1956. [Online]. Available: https://doi.org/10.1145/3292500.3330648
F.-M. De Rainville, F.-A. Fortin, M.-A. Gardner, M. Parizeau, and C. Gagn´e, “DEAP – enabling nimbler evolutions,” ACM SIGEVOlution, vol. 6, no. 2, pp. 17–26, 02 2014. [Online]. Available: https://doi.org/10.1145/2597453.2597455
S. Meisenbacher, M. Turowski, K. Phipps, M. R¨atz, D. M¨uller, V. Hagenmeyer, and R. Mikut, “Review of automated time series forecasting pipelines,” Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol. 12, no. 6, p. e1475, 2022. [Online]. Available: https://doi.org/10.1002/widm.1475
D. N. Papadopoulos, F. D. Javan, B. Najafi, A. H. Mamaghani, and F. Rinaldi, “Handling complete short-term data logging failure in smart buildings: Machine learning based forecasting pipelines with sliding-window training scheme,” Energy and Buildings, vol. 301, p. 113694, 2023. [Online]. Available: https://doi.org/10.1016/j.enbuild.2023.113694
T. P. Melo, J. O. Andrade, and K. S. Komati, “Impacto da ordem entre a seleção de características e a janela deslizante em séries temporais multivariadas na previsão do consumo de gás na pelotização de minério de ferro,” in Anais do XXV Congresso Brasileiro de Autom´atica (CBA 2024). Rio de Janeiro, RJ: SBA, Sociedade Brasileira de Automa¸c˜ao, 2024. [Online]. Available: https://www.sba.org.br/cba2024/papers/paper 9730.pdf
B. Dhariyal, T. Le Nguyen, and G. Ifrim, “Back to basics: A sanity check on modern time series classification algorithms,” in International Workshop on Advanced Analytics and Learning on Temporal Data. Springer, 2023, pp. 205–229. [Online]. Available: https://doi.org/10.1007/978-3-031-49896-1 14
M. R. Islam, A. A. Lima, S. C. Das, M. F. Mridha, A. R. Prodeep, and Y. Watanobe, “A comprehensive survey on the process, methods, evaluation, and challenges of feature selection,” IEEE Access, vol. 10, pp. 99 595–99 632, 2022. [Online]. Available: https://doi.org/10.1109/ACCESS.2022.3205618
M. Haddouchi and A. Berrado, “A survey and taxonomy of methods interpreting random forest models,” arXiv preprint arXiv:2407.12759, 2024. [Online]. Available: https://doi.org/10.48550/arXiv.2407.12759
H. J. Escalante, “Automated machine learning—a brief review at the end of the early years,” Automated Design of Machine Learning and Search Algorithms, pp. 11–28, 2021. [Online]. Available: https://doi.org/10.1007/978-3-030-72069-8 2
A. Truong, A. Walters, J. Goodsitt, K. Hines, C. B. Bruss, and R. Farivar, “Towards automated machine learning: Evaluation and comparison of automl approaches and tools,” in 2019 IEEE 31st international conference on tools with artificial intelligence (ICTAI). IEEE, 2019, pp. 1471–1479. [Online]. Available: https://doi.org/10.1109/ICTAI.2019.00209
A. Baldominos, Y. Saez, and P. Isasi, “On the automated, evolutionary design of neural networks: past, present, and future,” Neural Computing and Applications, p. 519–545, 2020. [Online]. Available: https://doi.org/10.1007/s00521-019-04160-6
B. Qu, P. Liao, and Y. Huang, “Outlier detection and forecasting for bridge health monitoring based on time series intervention analysis.” Structural Durability & Health Monitoring (SDHM), vol. 16, no. 4, 2022. [Online]. Available: https://doi.org/10.32604/sdhm.2022.021446
M. Haakman, L. Cruz, H. Huijgens, and A. Van Deursen, “AI lifecycle models need to be revised: An exploratory study in Fintech,” Empirical Software Engineering, vol. 26, no. 5, p. 95, 2021. [Online]. Available: https://doi.org/10.1007/s10664-021-09993-1
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Thadeu Pezzin Melo, Jefferson Oliveira Andrade, Karin Satie Komati
This work is licensed under a Creative Commons Attribution 4.0 International License.
CLEIej is supported by its home institution, CLEI, and by the contribution of the Latin American and international researchers community, and it does not apply any author charges whatsoever for submitting and publishing. Since its creation in 1998, all contents are made publicly accesibly. The current license being applied is a (CC)-BY license (effective October 2015; between 2011 and 2015 a (CC)-BY-NC license was used).
