Natural Gas Consumption Forecasting with Kolmogorov–Arnold Networks: A Comparison with MLP

Abstract

Natural gas remains a vital resource for meeting residential heating energy needs, particularly during the winter months. Accurate demand forecasting is essential for maintaining supply-demand balance, optimizing operational costs, and supporting effective energy management. In this study, the natural gas consumption prediction performance of Kolmogorov-Arnold Networks (KAN), a new neural network architecture, was compared with that of the basic model, Multi-Layer Perceptrons (MLP). Both models were trained and tested on the same dataset using monthly consumption data. While MLPs are diversified through the number of neurons, activation functions, and layer configuration, KAN models are configured by modifying B-spline parameters, grid size, and layer structure. The results show that the KAN model achieved the highest R2 value despite having fewer trained parameters. Although some versions of the MLP model yielded lower Mean Absolute Percentage Error (MAPE) values, they fell short of KAN in terms of overall fit. These findings demonstrate the superior ability of KAN to capture nonlinear patterns in energy demand forecasting, offering computational efficiency.

Keywords

• Kolmogorov-Arnold Networks (KANs)
• Natural Gas
• Demand Forecasting
• MLP
• Time series

References

1. K. Arslan, M. Akpınar, and M. Fatih Adak, “The detection of unaccounted natural gas consumption: A neural networks and subscriber-based solution,” Engineering Science and Technology, an International Journal, vol. 52, p. 101669, Apr. 2024, doi: 10.1016/j.jestch.2024.101669.
2. N. Wei et al., “Data complexity of daily natural gas consumption: Measurement and impact on forecasting performance,” Energy, vol. 238, 2022, doi: 10.1016/j.energy.2021.122090.
3. M. Akpinar, M. F. Adak, and N. Yumusak, “Forecasting natural gas consumption with hybrid neural networks — Artificial bee colony,” in 2016 2nd International Conference on Intelligent Energy and Power Systems (IEPS), IEEE, Jun. 2016, pp. 1–6. doi: 10.1109/IEPS.2016.7521852.
4. M. Akpinar and N. Yumusak, “Year ahead demand forecast of city natural gas using seasonal time series methods,” Energies, vol. 9, no. 9, 2016, doi: 10.3390/en9090727.
5. M. Akpınar and N. Yumusak, “Estimating household natural gas consumption with multiple regression: Effect of cycle,” in 2013 International Conference on Electronics, Computer and Computation, ICECCO 2013, 2013, doi: 10.1109/ICECCO.2013.6718260.
6. M. Akpınar and N. Yumusak, “Forecasting household natural gas consumption with ARIMA model: A case study of removing cycle,” in AICT 2013 - 7th International Conference on Application of Information and Communication Technologies, Conference Proceedings, 2013, doi: 10.1109/ICAICT.2013.6722753.
7. Q. Wang, S. Liu, and H. Yan, “The application of trigonometric grey prediction model to average per capita natural gas consumption of households in China,” GS, vol. 9, no. 1, pp. 19–30, Feb. 2019, doi: 10.1108/GS-08-2018-0033.
8. Z. Mi et al., “China’s Energy Consumption in the New Normal,” Earth’s Future, vol. 6, no. 7, 2018, doi: 10.1029/2018EF000840.

9. G. De and W. Gao, “Forecasting China’s natural gas consumption based on adaboost-particle swarm optimization-extreme learning machine integrated learning method,” Energies, vol. 11, no. 11, 2018, doi: 10.3390/en11112938.
10. X. Wang, D. Luo, J. Liu, W. Wang, and G. Jie, “Prediction of natural gas consumption in different regions of China using a hybrid MVO-NNGBM model,” Mathematical Problems in Engineering, vol. 2017, 2017, doi: 10.1155/2017/6045708.
11. A. S. Anđelković and D. Bajatović, “Integration of weather forecast and artificial intelligence for a short-term city-scale natural gas consumption prediction,” Journal of Cleaner Production, vol. 266, 2020, doi: 10.1016/j.jclepro.2020.122096.
12. W. Qiao, Z. Yang, Z. Kang, and Z. Pan, “Short-term natural gas consumption prediction based on Volterra adaptive filter and improved whale optimization algorithm,” Engineering Applications of Artificial Intelligence, vol. 87, p. 103323, Jan. 2020, doi: 10.1016/j.engappai.2019.103323.
13. Z. Liu et al., “KAN: Kolmogorov-Arnold Networks,” 2024, arXiv. doi: 10.48550/ARXIV.2404.19756.
14. C. J. Vaca-Rubio, L. Blanco, R. Pereira, and M. Caus, “Kolmogorov-Arnold Networks (KANs) for Time Series Analysis,” 2024, arXiv. doi: 10.48550/ARXIV.2405.08790.
15. M. H. Sulaiman, Z. Mustaffa, A. I. Mohamed, A. S. Samsudin, and M. I. Mohd Rashid, “Battery state of charge estimation for electric vehicle using Kolmogorov-Arnold networks,” Energy, vol. 311, p. 133417, Dec. 2024, doi: 10.1016/j.energy.2024.133417.
16. M. H. Sulaiman, Z. Mustaffa, M. S. Saealal, M. M. Saari, and A. Z. Ahmad, “Utilizing the Kolmogorov-Arnold Networks for chiller energy consumption prediction in commercial building,” Journal of Building Engineering, vol. 96, p. 110475, Nov. 2024, doi: 10.1016/j.jobe.2024.110475.
17. Y. Peng et al., “Predictive modeling of flexible EHD pumps using Kolmogorov–Arnold Networks,” Biomimetic Intelligence and Robotics, vol. 4, no. 4, p. 100184, Dec. 2024, doi: 10.1016/j.birob.2024.100184.
18. F. Granata, S. Zhu, and F. Di Nunno, “Advanced streamflow forecasting for Central European Rivers: The Cutting-Edge Kolmogorov-Arnold networks compared to Transformers,” Journal of Hydrology, vol. 645, p. 132175, Dec. 2024, doi: 10.1016/j.jhydrol.2024.132175.
19. IGDAS, “Monthly Natural Gas Consumption by District.” Accessed: Oct. 09, 2024. [Online]. Available: https://data.ibb.gov.tr/
20. X. Feng, G. Ma, S.-F. Su, C. Huang, M. K. Boswell, and P. Xue, “A multi-layer perceptron approach for accelerated wave forecasting in Lake Michigan,” Ocean Engineering, vol. 211, p. 107526, Sep. 2020, doi: 10.1016/j.oceaneng.2020.107526.
21. F. Taşpınar, N. Çelebi, and N. Tutkun, “Forecasting of daily natural gas consumption on regional basis in Turkey using various computational methods,” Energy and Buildings, vol. 56, pp. 23–31, Jan. 2013, doi: 10.1016/j.enbuild.2012.10.023.
22. B. Soldo, P. Potočnik, G. Šimunović, T. Šarić, and E. Govekar, “Improving the residential natural gas consumption forecasting models by using solar radiation,” Energy and Buildings, vol. 69, pp. 498–506, Feb. 2014, doi: 10.1016/j.enbuild.2013.11.032.
23. J. Szoplik, “Forecasting of natural gas consumption with artificial neural networks,” Energy, vol. 85, pp. 208–220, Jun. 2015, doi: 10.1016/j.energy.2015.03.084.
24. M. Akpinar, M. Adak, and N. Yumusak, “Day-Ahead Natural Gas Demand Forecasting Using Optimized ABC-Based Neural Network with Sliding Window Technique: The Case Study of Regional Basis in Turkey,” Energies, vol. 10, no. 6, p. 781, Jun. 2017, doi: 10.3390/en10060781.
25. Q. Wang and F. Jiang, “Integrating linear and nonlinear forecasting techniques based on grey theory and artificial intelligence to forecast shale gas monthly production in Pennsylvania and Texas of the United States,” Energy, vol. 178, pp. 781–803, Jul. 2019, doi: 10.1016/j.energy.2019.04.115.
26. E. Fabbiani, A. Marziali, and G. D. Nicolao, “Ensembling methods for countrywide short term forecasting of gas demand,” IJOGCT, vol. 26, no. 2, p. 184, 2021, doi: 10.1504/IJOGCT.2021.10035077.
27. R. Hribar, P. Potočnik, J. Šilc, and G. Papa, “A comparison of models for forecasting the residential natural gas demand of an urban area,” Energy, vol. 167, pp. 511–522, Jan. 2019, doi: 10.1016/j.energy.2018.10.175.
28. W. Qiao, K. Huang, M. Azimi, and S. Han, “A Novel Hybrid Prediction Model for Hourly Gas Consumption in Supply Side Based on Improved Whale Optimization Algorithm and Relevance Vector Machine,” IEEE Access, vol. 7, pp. 88218–88230, 2019, doi: 10.1109/ACCESS.2019.2918156.