Page 92 - 2023-Vol19-Issue2
P. 92
88 | Swide & Marhoon
is 0.00025, both of which decrease as the number of epochs [7] H. Cai, S. Shen, Q. Lin, X. Li, and H. Xiao, “Predict-
rises. Based on the aforementioned findings, we opt for the ing the energy consumption of residential buildings for
70-piece batch size with the highest accuracy and lowest loss. regional electricity supply-side and demand-side man-
agement,” IEEE Access, vol. 7, pp. 30386–30397, 2019.
V. CONCLUSION
[8] G. K. F. Tso and K. K. W. Yau, “Predicting electricity
In this study, two models for predicting and managing res- energy consumption: A comparison of regression analy-
idential load usage in smart grids are introduced. The first sis, decision tree and neural networks,” Energy, vol. 32,
model seeks to predict the load demand consumption using no. 9, pp. 1761–1768, 2007.
a deep learning (DL) model that can automatically train and
extract the model parameters from the historical data, utiliz- [9] M. Kankal, A. Akpinar, M. I. Komurcu, and T. s. ozsahin,
ing deep RNN, LSTM units, and accurately predict the load “Modeling and forecasting of turkey’s energy consump-
demand consumption. The other proposal is called random tion using socio-economic and demographic variables,”
forest, and when comparing deep learning and RF models, Appl. Energy, vol. 88, no. 5, pp. 1927–1939, 2011.
deep learning came out on top in terms of mean square error.
At the same time, RF was preferred based on the R2 Score be- [10] K. B. Debnath and M. Mourshed, “Forecasting methods
cause it had the highest accuracy (98%), which was achieved in energy planning models,” Renew. Sustain. Energy
when predicting electrical energy consumption. To improve Rev., vol. 88, pp. 297–325, 2018.
accuracy, future research will compare one deep learning
model and support vector machines. [11] H. Shi, M. Xu, and R. Li, “Deep learning for household
load forecasting-a novel pooling deep rnn,” IEEE Trans.
CONFLICT OF INTEREST Smart Grid, vol. 9, no. 5, pp. 5271–5280, 2018.
The authors have no conflict of relevant interest to this [12] A. Rahman, V. Srikumar, and A. D. Smith, “Predicting
article. electricity consumption for commercial and residential
buildings using deep recurrent neural networks,” Appl.
REFERENCES Energy, vol. 212, pp. 372–385, Feb. 2018.
[1] Y. W. Su, “Residential electricity demand in taiwan: [13] J. Nagi, K. S. Yap, F. Nagi, S. K. Tiong, and S. K.Ahmed,
Consumption behavior and rebound effect,” Energy Pol- “A computational intelligence scheme for the prediction
icy, vol. 124, no. June 2018, pp. 36–45, 2019. of the daily peak load,” Appl. Soft Comput. J., vol. 11,
no. 8, pp. 4773–4788, 2011.
[2] M. Shahbaz, S. Sarwar, W. Chen, and M. N. Malik,
“Dynamics of electricity consumption, oil price and [14] Y. Y. Cheng, P. P. K. Chan, and Z. W. Qiu, “Random
economic growth: Global perspective,” Energy Policy, forest based ensemble system for short term load fore-
vol. 108, no. 79532, pp. 256–270, 2017. casting,” Proc. - Int. Conf. Mach. Learn. Cybern., vol. 1,
pp. 52–56, 2012.
[3] I. Al-Kharsan, A. Marhoon, and J. Mahmood, “Fair
and balance demand response application in distribution [15] M. W. Ahmad, M. Mourshed, and Y. Rezgui, “Trees
networks,” Iraqi J. Electr. Electron. Eng., vol. sceeer, vs neurons: Comparison between random forest and
no. 3d, pp. 139–151, 2020. ann for high-resolution prediction of building energy
consumption,” Energy Build, vol. 147, pp. 77–89, 2017.
[4] S. Kulkarni, S. P. Simon, and K. Sundareswaran, “A
spiking neural network (snn) forecast engine for short- [16] D. Lee, S. Kang, and J. Shin, “Using deep learning
term electrical load forecasting,” Appl. Soft Comput. J., techniques to forecast environmental consumption level,”
vol. 13, no. 8, pp. 3628–3635, 2013. Sustain., vol. 9, no. 10, pp. 1–17, 2017.
[5] Y. Tian, J. Yu, and A. Zhao, “Predictive model of energy [17] C. Li, Z. Ding, D. Zhao, J. Yi, and G. Zhang, “Build-
consumption for office building by using improved gwo- ing energy consumption prediction: An extreme deep
bp,” Energy Reports, vol. 6, pp. 620–627, 2020. learning approach,” Energies, vol. 10, no. 10, pp. 1–20,
2017.
[6] S. G. Yoo and H. A. Myriam, “Predicting residential
electricity consumption using neural networks: A case [18] Z. Wang, Y. Wang, R. Zeng, R. S. Srinivasan, and
study,” J. Phys. Conf. Ser., vol. 1072, no. 1, 2018. S. Ahrentzen, “Random forest based hourly building
energy prediction,” Energy Build., vol. 171, pp. 11–25,
2018.
