Evaluation of Medium-Term Renewable Energy Generation Forecasting of EGAT Using Artificial Neural Networks and ARIMA
Keywords:
artificial neural networks, ARIMA, renewable energy generation forecasting, Electricity Generating Authority of Thailand (EGAT)Abstract
This study aims to develop and evaluate Neural Network and ARIMA models’ performance in forecasting electricity generation from renewable energy sources. The analysis used historical data on renewable energy generation from the Electricity Generating Authority of Thailand (EGAT). Two data sets were employed: the first data sets from the previous 3 months, 2 months, 1 month, and a month code, while the second data sets from the previous 4 months, 3 months, 2 months, 1 month, and a month code. The findings demonstrate that Neural Networks outperformed ARIMA models, achieving an MAPE of 3.3151%, compared to 7.4184% for ARIMA. This indicates that Neural Networks are better equipped to handle the inherent complexities of renewable energy generation data, making them more suitable for practical applications in electricity forecasting. These results suggest that the adoption of Neural Network models could enable EGAT to enhance the accuracy of its energy planning, minimize resource wastage, and optimize the utilization of renewable energy resources. Furthermore, this study underscores the potential for Neural Networks to advance EGAT’s capabilities in renewable energy management, supporting future innovations in this domain.
References
Abisoye, B. O., Sun, Y., & Zenghui, W. (2024). A survey of artificial intelligence methods for renewable energy forecasting: Methodologies and insights. Renewable Energy Focus, 48, 100529. https://doi.org/10.1016/j.ref.2023.100529
AlShafeey, M., & Csaki, C. (2024). Adaptive machine learning for forecasting in wind energy: A dynamic, multi-algorithmic approach for short and long-term predictions. Heliyon, 10, e34807. https://doi.org/10.1016/j.heliyon.2024.e34807
Banda, E., & Folly, K. A. (2007). Short Term Load Forecasting Using Artificial Neural Network. 2007 IEEE Lausanne Power Tech (pp. 108-112). Lausanne: IEEE
Dindem, P., Jirivibhakorn, S., & Keturksa, S. (2019). Long-term electricity demand forecasting for the EGAT system using neural networks. In Proceedings of the 42nd Electrical Engineering Conference (pp. 41-44). Bangkok: Mahidol University. (in Thai)
Energy Policy and Planning Office. (2024). Capacity, generation, peak, consumption, import, export and fuel used. Retrieved from https://www.eppo.go.th/index.php/th/energy-information/static-energy/static-electricity (in Thai)
Electricity Generating Authority of Thailand. (2024). Renewable energy: Background and development. Retrieved from https://www.egat.co.th/home/renewables (in Thai)
IBM Corp. (2024). IBM SPSS statistics (computer software). Retrieved from https://www.ibm.com/products/spss-statistics. (in Thai)
Jiriwibhakorn, S. (2011). Applications of neural networks in power system. Bangkok: Faculty of Engineering King Mongkut’s Institute of Technology Ladkrabang. (in Thai)
Jiriwibhakorn, S. (2022). Forecasting machines in power systems. Bangkok: Faculty of Engineering King Mongkut’s Institute of Technology Ladkrabang. (in Thai)
Katruksa, S. (2013). Facility energy sage modeling and medium term load forecasting with artificial Neural Networks And Adaptive Neuro-Fuzzy Inference Systems: An exploratory Study (Master’s thesis). King Mongkut’s Institute of Technology Ladkrabang. (in Thai)
Katruksa, S. (2020). Mid-term load forecasting based on artificial neural networks and adaptive Neuro-Fuzzy Inference Systems: An exploratory study (Doctoral dissertation). KingMongkut’s Institute of Technology Ladkrabang. (in Thai)
Katruksa, S., & Jiriwibhakorn, S. (2015). Facility energy usage modeling and medium term load forecasting with Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems. Kasetsart Engineering Journal, 28(91), 61-69. https://ph01.tci-thaijo.org/index.php/kuengj/article/view/79349/63422 (in Thai)
Katruksa, S., & Jiriwibhakorn, S. (2019). Application data for electricity load forecasting models. The 5th International Conference on Engineering, Applied Sciences and Technology (pp. 450-453). Luang Prabang: PDR.
Katruksa, S., & Jiriwibhakorn, S. (2020). Evaluation of mid-term load forecasting case study based on Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANNs). International Review of Electrical Engineering (I.R.E.E.), 15(4), 283-293. https://doi.org/10.15866/iree.v15i4.17766
Katruksa, S., & Kullimratchai, P. (2024). Application of Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems in The Electrical Power System. EAU Heritage Journal Science and Technology, 18(1), 1–19. https://he01.tci-thaijo.org/index.php/EAUHJSci/article/view/266469 (in Thai)
MathWorks. (2024). Matlab R2024a (Computer software). Retrieved from https://www.mathworks.com/products/matlab.html (in Thai)
Meesut, P. (2009). Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems. Management Information Systems. Bangkok: King Mongkut’s University of Technology North Bangkok. (in Thai)
Singh, S., Parmar, K. S., & Kumar, J. (2024). Development of multi-forecasting model using Monte Carlo simulation coupled with wavelet denoising-ARIMA model. Mathematics and Computers in Simulation, 2024, Related articles. https://doi.org/10.1016/j.matcom.2024.10.040.
Tarmanini, C., Sarma, N., Gezegin, C., & Ozgonenel, O. (2023). Short term load forecasting based on ARIMA and ANN approaches. Energy Reports, 9, 550–557. https://doi.org/10.1016/j.egyr.2023.01.060
Tripathi, P. K., Sen, R., & Upadhyay, S. K. (2021). A Bayes algorithm for model compatibility and comparison of ARMA (p, q) models. Statistics in Transition New Series, 22(2), 95–123. https://www.econstor.eu/bitstream/10419/236830/1/10.21307_stattrans-2021-018.pdf
Wang, G., Su, H., Mo, L., Yi, X., & Wu, P. (2024). Forecasting of soil respiration time series via clustered ARIMA. Computers and Electronics in Agriculture, 225, 109315. https://doi.org/10.1016/j.compag.2024.109315
Zhong, W., Zhai, D., Xu, W., Gong, W., Yan, C., Zhang, Y., & Qi, L. (2024). Accurate and efficient daily carbon emission forecasting based on improved ARIMA. Applied Energy, 376, 124232. https://doi.org/10.1016/j.apenergy.2024.124232
