# Hourly Power Consumption Forecasting Using RobustSTL and TCN

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

- We propose a forecasting model for single time series data regarding hourly power consumption utilizing RobustSTL and TCN;
- The study’s key contribution is the hybrid model of RobustSTL and TCN as the forecasting model;
- The proposed model can capture and understand the time series data despite containing dynamic patterns and burstiness;
- The experimental stage was performed based on real hourly power consumption and validated with the existing forecasting models.

## 2. Materials and Methods

#### 2.1. RobustSTL

Algorithm 1. RobustSTL method summary |

Input: y_{t}, parameter configuration |

Output: τ_{t}, s_{t}, r_{t} |

Step 1: Denoise the time series data using bilateral filtering, ${{y}^{\prime}}_{t}={\displaystyle {\sum}_{j\in J}{w}_{j}^{t}{y}_{j}}$ |

Step 2: Calculate the relative trend, ${\overline{\tau}}_{t}^{r}=\{\begin{array}{c}0,\\ {\displaystyle {\sum}_{i=2}^{t}\nabla {\widehat{\tau}}_{i}},\end{array}\begin{array}{c}t=1\\ t\ge 2\end{array}$, ${{y}^{\u2033}}_{t}={{y}^{\prime}}_{t}-{\overline{\tau}}_{t}^{r}$ |

Step 3: Calculate the seasonality using non-local seasonal filtering, ${\overline{s}}_{t}={\displaystyle {\sum}_{({t}^{\prime},j)\in \phi}{w}_{({t}^{\prime},j)}^{t}{{y}^{\u2033}}_{j}}$ |

Step 4: Adjust the trend, seasonality, and remainder components |

${\tau}_{t}={\overline{\tau}}_{t}^{r}+{\tau}_{1}$, ${s}_{t}={\overline{s}}_{t}-{\tau}_{1}$,${r}_{t}={y}_{t}-{s}_{t}-{\tau}_{t}$ |

#### 2.2. TCN

#### 2.3. Evaluation Metrics

## 3. Results and Discussion

#### 3.1. Data Preparation

#### 3.2. Experimental Results

## 4. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Zheng, S.; Huang, G.; Zhou, X.; Zhu, X. Climate-change impacts on electricity demands at a metropolitan scale: A case study of Guangzhou, China. App. Energy
**2020**, 261, 114295. [Google Scholar] [CrossRef] - Albuquerque, P.C.; Cajueiro, D.O.; Rossi, M.D.C. Machine learning models for forecasting power electricity consumption using a high dimensional dataset. Expert Sys. App.
**2022**, 187, 115917. [Google Scholar] [CrossRef] - Lee, J.; Cho, Y. National-scale electricity peak load forecasting: Traditional, machine learning, or hybrid model. Energy
**2022**, 239, 122366. [Google Scholar] [CrossRef] - Petrosanu, D.M. Designing, developing and validating a forecasting method for the month ahead hourly electricity consumption in the case of medium industrial consumers. Processes
**2019**, 7, 310. [Google Scholar] [CrossRef][Green Version] - Sulandari, W.; Subanar; Lee, M.H.; Rodrigues, P.C. Indonesian electricity load forecasting using singular spectrum analysis, fuzzy systems and neural networks. Energy
**2020**, 190, 116408. [Google Scholar] [CrossRef] - Fan, M.; Hu, Y.; Zhang, X.; Yin, H.; Yang, Q.; Fan, L. Short-term load forecasting for distribution network using decomposition with ensemble prediction. In Proceedings of the Chinese Automation Congress (CAC), Hangzhou, China, 22–24 November 2019. [Google Scholar]
- Tian, Y.J.; Wen, M.; Li, J.G. A short-term electricity forecasting scheme based on combined GRU model with STL decomposition. In Proceedings of the IOP Conference Series: Earth and Environmental Science, Xi’an, China, 3–6 November 2020. [Google Scholar]
- Mendez-Jimenez, I.; Cardenas-Montes, M. Time series decomposition for improving the forecasting performance of convolutional neural networks. In Proceedings of the Advances in Artificial Intelligence, Granada, Spain, 23–26 October 2018. [Google Scholar]
- Wen, Q.; Gao, J.; Song, X.; Sun, L.; Xu, H.; Zhu, S. RobustSTL: A robust seasonal-trend decomposition algorithm for long time series. In Proceedings of the 33th AAAI Conference on Artificial Intelligence (AAAI-19), Honolulu, HI, USA, 27 January–1 February 2019. [Google Scholar]
- Jiang, F.; Zhang, C.; Sun, S.; Sun, J. Forecasting hourly PM based on deep temporal convolutional neural network and decomposition method. App. Soft Comp.
**2021**, 113, 107988. [Google Scholar] [CrossRef] - Wang, Z.; Tian, J.; Fang, H.; Chen, L.; Qin, J. LightLog: A lightweight temporal convolutional network for log anomaly detection on the edge. Comp. Net.
**2022**, 203, 108616. [Google Scholar] [CrossRef] - Cleveland, R.B.; Cleveland, W.S.; McRae, J.E.; Terpenning, I. STL: A seasonal-trend decomposition procedure based on Loess. J. Off. Stat.
**1990**, 6, 3–73. [Google Scholar] - Bai, S.; Kolter, J.Z.; Koltun, V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv
**2018**, arXiv:1803.01271. [Google Scholar] - Yuan, X.; Qi, S.; Wang, Y.; Wang, K.; Yang, C.; Ye, L. Quality variable prediction for nonlinear dynamic industrial processes based on temporal convolutional networks. IEEE Sens. J.
**2021**, 21, 20493–20503. [Google Scholar] [CrossRef] - Yin, L.; Xie, J. Multi-feature-scale fusion temporal convolution networks for metal temperature forecasting of ultra-supercritical coal-fired power plant reheater tubes. Energy
**2022**, 238, 121657. [Google Scholar] [CrossRef] - Li, J.; Wu, Y.; Li, Y.; Xiang, J.; Zheng, B. The temperature prediction of hydro-generating units based on temporal convolutional network and recurrent neural network. In Proceedings of the 40th Chinese Control Conference (CCC), Shanghai, China, 26–28 July 2021. [Google Scholar]
- Katarya, R.; Verma, O.P. Effectual recommendations using artificial algae algorithm and fuzzy c-mean. Swarm Evol. Comput.
**2017**, 36, 52–61. [Google Scholar] [CrossRef] - Deng, X.; Liu, Q.; Deng, Y.; Mahdevan, S. An improved method to construct basic probability assignment based on the confusion matrix for classification problem. Inf. Sci.
**2016**, 340–341, 250–261. [Google Scholar] [CrossRef] - Hourly Power Consumption of Turkey (2016–2020). Available online: https://www.kaggle.com/datasets/hgultekin/hourly-power-consumption-of-turkey-20162020 (accessed on 1 November 2021).
- –Namina, S.S.; Tavakoli, N.; Namin, A.S. A Comparison of ARIMA and LSTM in Forecasting Time Series. In Proceedings of the 17th IEEE International Conference on Machine Learning and Applications (ICMLA), Orlando, FL, USA, 17–20 December 2018. [Google Scholar]
- Liu, B.; Fu, C.; Bielefield, A.; Liu, Y.Q. Forecasting of Chinese Primary Energy Consumption in 2021 with GRU Artificial Neural Network. Energies
**2017**, 10, 1453. [Google Scholar] [CrossRef] - Yang, S.; Deng, Z.; Li, X.; Zheng, C.; Xi, L.; Zhuang, J.; Zhang, Z.; Zhang, Z. A novel hybrid model based on STL decomposition and one-dimensional convolutional neural networks with positional encoding for significant wave height forecast. Renew. Energy
**2021**, 173, 531–543. [Google Scholar] [CrossRef]

Model | MAPE (%) | MAE | RMSE |
---|---|---|---|

LSTM | 3.56 | 0.93 | 1.37 |

GRU | 3.51 | 0.94 | 1.29 |

STL-GRU | 2.34 | 0.66 | 0.88 |

RobustSTL-CNN | 1.95 | 0.58 | 0.85 |

The proposed model | 1.89 | 0.55 | 0.81 |

Model | Precision | Recall | F1-Score |
---|---|---|---|

LSTM | 0.65 | 0.64 | 0.64 |

GRU | 0.61 | 0.63 | 0.62 |

STL-GRU | 0.60 | 0.61 | 0.60 |

RobustSTL-CNN | 0.63 | 0.62 | 0.62 |

The proposed model | 0.70 | 0.70 | 0.70 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Lin, C.-H.; Nuha, U.; Lin, G.-Z.; Lee, T.-F. Hourly Power Consumption Forecasting Using RobustSTL and TCN. *Appl. Sci.* **2022**, *12*, 4331.
https://doi.org/10.3390/app12094331

**AMA Style**

Lin C-H, Nuha U, Lin G-Z, Lee T-F. Hourly Power Consumption Forecasting Using RobustSTL and TCN. *Applied Sciences*. 2022; 12(9):4331.
https://doi.org/10.3390/app12094331

**Chicago/Turabian Style**

Lin, Chih-Hsueh, Ulin Nuha, Guang-Zhi Lin, and Tsair-Fwu Lee. 2022. "Hourly Power Consumption Forecasting Using RobustSTL and TCN" *Applied Sciences* 12, no. 9: 4331.
https://doi.org/10.3390/app12094331