Hybrid Optimized Machine Learning Models for Forecasting Electricity Demand and Generation: A Comparative Analysis

Liandian Jiang, Pengfei Huang, Junyang Tian, Haiyong Li, Bin Liu, Shan Liang, Jian Wang

Abstract


Increasingly complicated and volatile electricity demand considerably complicates the job of operators and planners of energy systems. For enhancing grid reliability and efficiency in resource allocation, realistic forecasting is thus in high demand regarding overall electricity demand and peak generation capacity. The study will introduce a new approach by adopting ML techniques, namely CatBoost and support vector regression, for predicting electricity demand and maximum generation capacity. Optimization algorithms for metaheuristics, such as phasor particle swarm optimizer and chaos game optimizer, are used to improve model performance. To this end, four hybrid models were developed: namely, CatBoost-CGO, CatBoost-PPSO, SVR-CGO, and SVR-PPSO. The performance was evaluated using statistical indices such as MAE, RMSE, MAPE, and R². Among the models, CatBoost-PPSO demonstrated the highest accuracy with R² = 0.955 for total demand and R² = 0.958 for maximum generation in the test set, outperforming other combinations. These results validate the effectiveness of hybrid optimization in improving prediction accuracy.


Full Text:

PDF

References


Y. C. Hu, “Electricity consumption prediction using a neural-network-based grey forecasting approach,” Journal of the Operational Research Society, 68(10):1259–1264, Oct. 2017, Doi: 10.1057/s41274-016-0150-y.

https://doi.org/10.1057/s41274-016-0150-y

L. Ekonomou, “Greek long-term energy consumption prediction using artificial neural networks,” Energy, 35, (2): 512–517, 2010, Doi: 10.1016/j.energy.2009.10.018.

K. Li, C. Hu, G. Liu, and W. Xue, “Building’s electricity consumption prediction using optimized artificial neural networks and principal component analysis,” Energy Build, 108, 106–113, Dec. 2015, Doi: 10.1016/j.enbuild.2015.09.002.

W. Zhang and J. Yang, “Forecasting natural gas consumption in China by Bayesian Model Averaging,” Energy Reports, 1, 216–220, Nov. 2015, Doi: 10.1016/j.egyr.2015.11.001.

N. Fumo and M. A. Rafe Biswas, “Regression analysis for prediction of residential energy consumption,” 2015, Elsevier Ltd. Doi: 10.1016/j.rser.2015.03.035.

S. Barak and S. S. Sadegh, “Forecasting energy consumption using ensemble ARIMA-ANFIS hybrid algorithm,” International Journal of Electrical Power and Energy Systems, 82, 92–104, Nov. 2016, Doi: 10.1016/j.ijepes.2016.03.012.

J. Hwang, D. Suh, and M. O. Otto, “Forecasting electricity consumption in commercial buildings using a machine learning approach,” Energies (Basel), 13(22) Nov. 2020, Doi: 10.3390/en13225885.

R. Wang, S. Lu, and W. Feng, “A novel improved model for building energy consumption prediction based on model integration,” Appl Energy, 262, Mar. 2020, Doi: 10.1016/j.apenergy.2020.114561.

Z. Lin, L. Cheng, and G. Huang, “Electricity consumption prediction based on LSTM with attention mechanism,” IEEJ Transactions on Electrical and Electronic Engineering, 15(4):556–562, Apr. 2020, Doi: 10.1002/tee.23088.

G. Vijendar Reddy, L. J. Aitha, C. Poojitha, A. Naga Shreya, D. Krithika Reddy, and G. Sai Meghana, “Electricity Consumption Prediction Using Machine Learning,” in E3S Web of Conferences, EDP Sciences, Jun. 2023. 10.1051/e3sconf/202339101048.

R. Olu-Ajayi and H. Alaka, “Building energy consumption prediction using deep learning.”

A. V. Dorogush, V. Ershov, and A. Gulin, “CatBoost: gradient boosting with categorical features support,” Oct. 2018, [Online]. Available: http://arxiv.org/abs/1810.11363

G. Huang et al., “Evaluation of CatBoost method for prediction of reference evapotranspiration in humid regions,” J Hydrol (Amst), 574, 1029–1041, Jul. 2019, Doi: 10.1016/j.jhydrol.2019.04.085.

H. Drucker·, C. J. C. Burges, L. Kaufman, A. Smola··, and V. Vapoik, “Support Vector Regression Machines.” https://doi.org/10.1007/978-0-387-30162-4_415

B. Üstün, W. J. Melssen, and L. M. C. Buydens, “Facilitating the application of Support Vector Regression by using a universal Pearson VII function-based kernel,” Chemometrics and Intelligent Laboratory Systems, 81(1):29–40, Mar. 2006, Doi: 10.1016/j.chemolab.2005.09.003.

S. Talatahari and M. Azizi, “Chaos Game Optimization: a novel metaheuristic algorithm,” Artif Intell Rev, 54(2): 917–1004, Feb. 2021, Doi: 10.1007/s10462-020-09867-w.

A. Ramadan, S. Kamel, M. M. Hussein, and M. H. Hassan, “A new application of chaos game optimization algorithm for parameters extraction of three diode photovoltaic model,” IEEE Access, 9, 51582–51594, 2021, Doi: 10.1109/ACCESS.2021.3069939.

J. Kennedy, R. Eberhart, and bls gov, “Particle Swarm Optimization.” DOI:10.1109/ICNN.1995.488968

D. N. Wilke, M. Khajehzadeh, and Q. Bai, “Analysis of Particle Swarm Optimization Algorithm Swarm Int elligence based Soft Comput ing Techniques for the Solut ions to Mult iobject ive Op… hifza afaq A Survey of the St at e of the Art in Part icle Swarm Opt imizat ion A Review of Populat ion-based Met a-Heurist ic Algorit hms Analysis of Particle Swarm Optimization Algorithm.” https://doi.org/10.1007/s11831-021-09694-4

M. Ghasemi, E. Akbari, A. Rahimnejad, S. E. Razavi, S. Ghavidel, and L. Li, “Phasor particle swarm optimization: a simple and efficient variant of PSO,” Soft comput, 23(19): 9701–9718, Oct. 2019, doi: 10.1007/s00500-018-3536-8. https://doi.org/10.1007/s00500-018-3536-8

khatsu Siseyiekuo, Srivastava Abhishek, and Kumar Das Dushimanta, “An Adaptive Phasor Particle Swarm Optimization to Solve Economic Load Dispatch and Combined Emission Economic Load Dispatch Problem,” 2019. DOI:10.1109/UPCON47278.2019.8980245

F. Pianosi and T. Wagener, “A simple and efficient method for global sensitivity analysis based oncumulative distribution functions,” Environmental Modelling and Software, 67, 1–11, May 2015, Doi: 10.1016/j.envsoft.2015.01.004.

A. Puy, S. Lo Piano, and A. Saltelli, “A sensitivity analysis of the PAWN sensitivity index,” Environmental Modelling and Software, 127, May 2020, Doi: 10.1016/j.envsoft.2020.104679.

I. Salehin, S. M. Noman, and M. M. Hasan, “Electricity energy dataset ‘BanE-16’: Analysis of peak energy demand with environmental variables for machine learning forecasting,” Data Brief, 52, Feb. 2024, Doi: 10.1016/j.dib.2023.109967.




DOI: https://doi.org/10.31449/inf.v49i18.7737

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.