Smart Media Journal (스마트미디어저널)

THE KOREAN INSTITUTE OF SMART MEDIA (한국스마트미디어학회)
권호 표지
DOI QR코드

DOI QR Code

Short-and Mid-term Power Consumption Forecasting using Prophet and GRU

Prophet와 GRU을 이용하여 단중기 전력소비량 예측

  • 손남례 (전남대학교 소프트웨어중심대학사업단) ;
  • 강은주 (호남대학교 AI융합대학 정보통신공학과)
  • Received : 2023.07.25
  • Accepted : 2023.11.06
  • Published : 2023.12.29
Download PDF
⟨ Previous Next ⟩

Abstract

The building energy management system (BEMS), a system designed to efficiently manage energy production and consumption, aims to address the variable nature of power consumption within buildings due to their physical characteristics, necessitating stable power supply. In this context, accurate prediction of building energy consumption becomes crucial for ensuring reliable power delivery. Recent research has explored various approaches, including time series analysis, statistical analysis, and artificial intelligence, to predict power consumption. This paper analyzes the strengths and weaknesses of the Prophet model, choosing to utilize its advantages such as growth, seasonality, and holiday patterns, while also addressing its limitations related to data complexity and external variables like climatic data. To overcome these challenges, the paper proposes an algorithm that combines the Prophet model's strengths with the gated recurrent unit (GRU) to forecast short-term (2 days) and medium-term (7 days, 15 days, 30 days) building energy consumption. Experimental results demonstrate the superior performance of the proposed approach compared to conventional GRU and Prophet models.

빌딩에너지관리시스템(BEMS: Building Energy Management System)은 생산 및 소비되는 에너지를 효율적으로 관리하는 시스템이다. 그러나 건물 내 전력소비는 물리적인 특성상으로 인해 생산 및 소비가 일정하지 않아 안정적인 전력 공급이 필수적이다. 이에 따라 건물의 안정적인 전력 공급을 위해서는 정확한 건물 내 전력 소비 예측이 중요하다. 최근에는 시계열분석, 통계분석, 인공지능 등 다양한 방법을 이용하여 전력소비예측에 관한 연구가 진행되고 있다. 본 논문은 Prophet 모델의 장점과 단점을 분석하여 장점인 growth, seasonality, holidays를 선택하였고, Prophet 모델의 단점인 데이터의 복잡성과 외부변수(기후 데이터)의 제한성을 해결하기 위하여 GRU을 조합하여 단기(2일) 및 중기(7일, 15일, 30일) 전력소비량 예측 알고리즘을 제안한다. 실험결과, 제안한 방법은 기존 GRU 및 Prophet 모델보다 성능이 우수하였다.

Keywords

Acknowledgement

본 연구는 과학기술정보통신부 및 정보통신기획평가원의 소프트웨어중심대학사업의 연구결과 수행되었습니다.(2021-0-01409)

References

  1. Zhang, Z., Zou, P., Song, W., and Xiao, F., "Data-driven approaches for building energy management systems: A review," Renewable and Sustainable Energy Reviews, No. 104, pp. 144-158, 2019. 
  2. Deyslen M.H., Luis H.C., Angle L.Z., Oscar D., Felix S.G., "A Review of Strategies for Building Energy Management System: Model Predictive Control, Demand Side Management, Optimization, and Fault Detect & Diagnosis," Journal of Building Engineering, Jul. 2020. 
  3. Feng, Y., "Study on Medium and Long Term Power Load Forecasting Based on Combination Forecasting Model," Chem. Eng. Trans. 2015, No. 51, pp. 859-864, 2016. 
  4. Xue, B. and Keng, J., "Dynamic transverse correction method of middle and long term energy forecasting based on statistic of forecasting errors," In Proceedings of the Conference on Power and Energy IPEC, Ho Chi Minh City, Vietnam, pp. 253-256, Dec. 2012. 
  5. R.G. Brown, "Smoothing Forecasting and Prediction of Discrete Time Series," vol. 127, no. 2, pp. 292-293 Jan. 1964.  https://doi.org/10.2307/2344012
  6. Y. Ohtsuka, T, Oga, and K. Kakamu, "Forecasting electricity demand in Japan: A Bayesian spatial autoregressive ARMA approach," Comp. Stat. Data Anal., No. 54, pp. 2721-2735, Nov. 2010. 
  7. C.E. Holt, "Forecasting Seasonal and Trends by Exponentially Weighted Average," International Journal of Forecasting, vol. 20, no. 1, pp. 5-10, 2004.  https://doi.org/10.1016/j.ijforecast.2003.09.015
  8. S.A. Kalogirou, C.C. Neocleous, and C.C. Schizas, "Building heating load estimation using artificial neural networks," In Proceedings of the 17th International Conference on Parallel Architectures and Compilation Techniques, pp. 10-14, San Francisco, CA, USA, Nov. 1997. 
  9. A. Bagnasco, F. Fresi, M. Saviozzi, F. Silvestro, and A. Vinci, "Electrical consumption forecasting in hospital facilities: An application case," Energy Build, No. 103, pp. 261-270, 2015. 
  10. F. Gers, J. Schmidhuber, and F. Cummins, "Learning to Forget: Continual Prediction with LSTM," In Proceedings of the 9th International Conference on Artificial Neural Networks, pp. 850-855, Edinburgh, UK, Sep. 1999. 
  11. P. Valenzuela, J. Gorricho, and de la Iglesia, "Automatic model and feature selection for time series forecasting: Achieving good performance and interpretability," Information Sciences, No. 423, pp. 157-174, 2018. 
  12. P. Sanguansat, and N. Klomjit, "A comparative study of machine learning techniques for short-term load forecasting," Energies, Vol. 12, No. 20, 2019. 
  13. G. Fung, and E. Shih, "Evaluating the Forecasting Performance of Facebook's Prophet Model for Time Series Data," Journal of Open Source Software, Vol. 4, No. 43, Sep. 2020. 
  14. Tasarruf Bashir and Chen Haoyong, "Short term electricity load forecasting using hybrid prophet-LSTM model optimized by BPNN," Energy Reports, Vol. 8, pp. 1678-1686, Nov. 2022  https://doi.org/10.1016/j.egyr.2021.12.067
  15. Pin Li and Jin-suo Zhang, " A New Hybrid Method for China's Energy Supply Security Forecasting Based on ARIMA and XGBoost," Energies, Vol. 11, no. 7, Jun. 2018. 
  16. Yuanhua Chen, Muhammad S.B., Muhammad, A., and Dingtian Xiao, "Evaluation of Machine Learning Models for Smart Grid Parameters: Performance Analysis of ARIMA and Bi-LSTM," Sustaiability, Vol. 15, no. 11, May 2023. 
  17. S. J. Taylor and B. Letham, "Prophet: forecasting at scale," Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1389-1397, 2017. 
  18. S.J Taylor and B. Letham, "Forecasing at Scale," The American Statistician, Taylor & Francis Journals, Vol. 72, no. 1, pp. 37-45, 2018. 
  19. K. Cho, Van Merrienboer, B., Bahdanau, D., and Y. Bengio, "On the properties of neural machine translation: Encoder-decoder approaches," arXiv:1409.1259, 2014. 
  20. Namrye son, "Comparison of the Deep Learning Performance for Short-Term Power Load Forecasting," sustainability, Vol. 13, No. 22, 2021.