Journal of Power Electronics and Power Systems Review Article
LightGBM-Based Grid Demand Estimation and Prediction for Intelligent Electricity Load Forecasting
Abstract
Correctly predicting electricity demand in emerging energy markets like India is the most effective way to maintain grid stability. This is particularly important. This article offers a comprehensive data- based structure that captures both the broad temporal dynamics and regional diversity. This technique utilizes a large-scale multi-dimensional feature space that includes temporal indicators, autoregressive lag terms, sliding-window rolling statistics, and state-level administrative markers. Our systematic benchmarking approach involves the use of four predictive architectures: XGBoost, CatBoost, LightGBM, and Gated Recurrent Units. Within the context of this engineered tabular paradigm, experiments reveal that gradient boosting models are far more effective than deep learning sequence models. LightGBM serves as the superior model, resulting in an R 2 of 99.56% and a MAPE of 4.93%. These results support the notion that utilizing extensive feature engineering with optimized decision trees can create a highly accurate, scalable, and operationally reliable approach to forecasting energy demand at governmental level. Moreover, the suggested method can be used to develop smart electricity load forecasting, which combines historical load data with temporal and geographical data. Chronological cross-validation is used to assess model performance, but maintains the temporal order of the observations, which is beneficial for forecasting applications. Comparisons of several models also emphasize the need to choose appropriate algorithms for grid demand estimation. The results highlight the promise of LightGBM forecasting for aiding data-driven electricity planning, grid management, and smart grid analytics in India.
Keywords
References (16)
- Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu TY. Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems. 2017;30.
- Friedman JH. Greedy function approximation: a gradient boosting machine. Annals of statistics. 2001 Oct 1:1189-232.
- Saxena A, Gupta A, Tiwari N. Design and Implementation of Adaptive and Artificial Intelligence Controller for Brushless Motor Drive Electric Vehicle. SAE International Journal of Electrified Vehicles. 2024 Jan 1;13(1):37-48.
- Patel S, Yadav S, Tiwari N. Design and optimization of PI controller for multiple-output wireless electric vehicle charger. Recent Advances in Electrical & Electronic Engineering. 2023;16(6):600-610. doi:10.2174/2352096516666230417084431.
- Tiwari N, Yadav S, Arya SR. PI gain optimisation and artificial intelligence based direct torque control of induction motor equipped electric vehicle drives. Int J Electr Hybrid Veh. 2023;15(2):151-182. doi:10.1504/IJEHV.2023.132034
- Breiman L. Random forests. Machine Learning. 2001;45(1):5–32. doi:10.1023/A:1010933404324.
- Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016. p. 785–794. doi:10.1145/2939672.2939785.
- Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A. CatBoost: Unbiased boosting with categorical features. Adv Neural Inf Process Syst. 2018;31:6639–6649. doi:10.48550/arXiv.1706.09516.
- Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine learning in Python. J Mach Learn Res. 2011;12:2825–2830. doi:10.5555/1953048.2078195.
- Attia PM, Grover A, Jin N, Severson KA, Markov TM, Liao YH, et al. Closed-loop optimization of fast- charging protocols for batteries with machine learning. Nature. 2020;578(7795):397–402. doi:10.1038/s41586-020-1994-5.
- Liu C, Wang Y, Chen Z. Degradation model and cycle life prediction for lithium-ion battery used in hybrid energy storage system. Energy. 2019;166:796–806. doi:10.1016/j.energy.2018.10.131.
- Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997;9(8):1735–1780. doi:10.1162/neco.1997.9.8.1735.
- Cho K, van Merriënboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, et al. Learning phrase representations using RNN encoder–decoder for statistical machine translation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2014. p. 1724–1734. doi:10.3115/v1/D14-1179.
- Chung J, Gulcehre C, Cho K, Bengio Y. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555. 2014. doi:10.48550/arXiv.1412.3555.
- Severson KA, Attia PM, Jin N, Perkins N, Jiang B, Yang Z, et al. Data-driven prediction of battery cycle life before capacity degradation. Nat Energy. 2019;4(5):383–391. doi:10.1038/s41526-019-0655-6.
- Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;30:4765–4774. doi:10.48550/arXiv.1705.07874.