journal article·2022-04-25Open access

Hourly Power Consumption Forecasting Using RobustSTL and TCN

Chih-Hsueh Lin, Ulin Nuha, Guang-Zhi Lin, Tsair-Fwu Lee

Applied Sciences· Volume 12· Issue 9

Abstract

Power consumption forecasting is a crucial need for power management to achieve sustainable energy. The power demand is increasing over time, while the forecasting of power consumption possesses challenges with nonlinearity patterns and various noise in the datasets. To this end, this paper proposes the RobustSTL and temporal convolutional network (TCN) model to forecast hourly power consumption. Through the RobustSTL, instead of standard STL, this decomposition method can extract time series data despite containing dynamic patterns, various noise, and burstiness. The trend, seasonality, and remainder components obtained from the decomposition operation can enhance prediction accuracy by providing significant information from the dataset. These components are then used as input for the TCN model applying deep learning for forecasting. TCN employing dilated causal convolutions and residual blocks to extract long-term data patterns outperforms recurrent networks in time series forecasting studies. To assess the proposed model, this paper conducts a comparison experiment between the proposed model and counterpart models. The result shows that the proposed model can grasp the rules of historical time series data related to hourly power consumption. Our proposed model overcomes the counterpart schemes in MAPE, MAE, and RMSE metrics. Additionally, the proposed model obtains the best results in precision, recall, and F1-score values. The result also indicates that the predicted data can fit the pattern of the actual data.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Energy Load and Power Forecasting · Air Quality Monitoring and Forecasting · Forecasting Techniques and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Electronic Engineering, National Kaohsiung University of Science and Technology, Kaohsiung 80778, Taiwan

Department of Tactical Control Air Traffic Control & Meteorology, Air Force Institute of Technology, Kaohsiung 82047, Taiwan

PhD Program in Biomedical Engineering, Kaohsiung Medical University, Kaohsiung 80708, Taiwan

Funding

Ministry of Science and Technology, Taiwan

References (22)

Show source-provided references
1. Climate-change impacts on electricity demands at a metropolitan scale: A case study of Guangzhou, China10.1016/j.apenergy.2019.114295 ↗
2. Machine learning models for forecasting power electricity consumption using a high dimensional dataset10.1016/j.eswa.2021.115917 ↗
3. National-scale electricity peak load forecasting: Traditional, machine learning, or hybrid model10.1016/j.energy.2021.122366 ↗
4. Petrosanu, D.M. (2019). Designing, developing and validating a forecasting method for the month ahead hourly electricity consumption in the case of medium industrial consumers. Processes, 7.10.3390/pr7050310 ↗
5. Indonesian electricity load forecasting using singular spectrum analysis, fuzzy systems and neural networks10.1016/j.energy.2019.116408 ↗
6. Fan, M., Hu, Y., Zhang, X., Yin, H., Yang, Q., and Fan, L. (2019, January 22–24). Short-term load forecasting for distribution network using decomposition with ensemble prediction. Proceedings of the Chinese Automation Congress (CAC), Hangzhou, China.10.1109/CAC48633.2019.8997169 ↗
7. Tian, Y.J., Wen, M., and Li, J.G. (2020, January 3–6). A short-term electricity forecasting scheme based on combined GRU model with STL decomposition. Proceedings of the IOP Conference Series: Earth and Environmental Science, Xi’an, China.
8. Mendez-Jimenez, I., and Cardenas-Montes, M. (2018, January 23–26). Time series decomposition for improving the forecasting performance of convolutional neural networks. Proceedings of the Advances in Artificial Intelligence, Granada, Spain.10.1007/978-3-030-00374-6_9 ↗
9. Wen, Q., Gao, J., Song, X., Sun, L., Xu, H., and Zhu, S. (February, January 27). RobustSTL: A robust seasonal-trend decomposition algorithm for long time series. Proceedings of the 33th AAAI Conference on Artificial Intelligence (AAAI-19), Honolulu, HI, USA.10.1609/aaai.v33i01.33015409 ↗
10. Forecasting hourly PM based on deep temporal convolutional neural network and decomposition method10.1016/j.asoc.2021.107988 ↗
11. LightLog: A lightweight temporal convolutional network for log anomaly detection on the edge10.1016/j.comnet.2021.108616 ↗
12. STL: A seasonal-trend decomposition procedure based on Loess
13. Bai, S., Kolter, J.Z., and Koltun, V. (2018). An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv.
14. Quality variable prediction for nonlinear dynamic industrial processes based on temporal convolutional networks10.1109/JSEN.2021.3096215 ↗
15. Multi-feature-scale fusion temporal convolution networks for metal temperature forecasting of ultra-supercritical coal-fired power plant reheater tubes10.1016/j.energy.2021.121657 ↗
16. Li, J., Wu, Y., Li, Y., Xiang, J., and Zheng, B. (2021, January 26–28). The temperature prediction of hydro-generating units based on temporal convolutional network and recurrent neural network. Proceedings of the 40th Chinese Control Conference (CCC), Shanghai, China.10.23919/CCC52363.2021.9549853 ↗
17. Effectual recommendations using artificial algae algorithm and fuzzy c-mean10.1016/j.swevo.2017.04.004 ↗
18. An improved method to construct basic probability assignment based on the confusion matrix for classification problem10.1016/j.ins.2016.01.033 ↗
19. (2021, November 01). Hourly Power Consumption of Turkey (2016–2020). Available online: https://www.kaggle.com/datasets/hgultekin/hourly-power-consumption-of-turkey-20162020.
20. –Namina, S.S., Tavakoli, N., and Namin, A.S. (2018, January 17–20). A Comparison of ARIMA and LSTM in Forecasting Time Series. Proceedings of the 17th IEEE International Conference on Machine Learning and Applications (ICMLA), Orlando, FL, USA.
21. Liu, B., Fu, C., Bielefield, A., and Liu, Y.Q. (2017). Forecasting of Chinese Primary Energy Consumption in 2021 with GRU Artificial Neural Network. Energies, 10.10.3390/en10101453 ↗
22. A novel hybrid model based on STL decomposition and one-dimensional convolutional neural networks with positional encoding for significant wave height forecast10.1016/j.renene.2021.04.010 ↗