Journal of the Korea Society of Computer and Information (한국컴퓨터정보학회논문지)

Korean Society of Computer Information (한국컴퓨터정보학회)
권호 표지
DOI QR코드

DOI QR Code

Enhancing Summer Electricity Demand Forecasting Using Fourier Transform-Based Time Variables

  • Received : 2024.09.20
  • Accepted : 2024.10.24
  • Published : 2024.11.29
Download PDF
⟨ Previous Next ⟩

Abstract

In the summer, when the cooling load rises due to high temperatures, the hourly demand increases during the day and is relatively less at night compared to the day. These characteristics are considered important information in predicting summer electricity demand. However, if time information is simply expressed as a dummy variable, the model simply recognizes differences between time zones rather than learning changes in time. In this study, we would like to approach this problem by using a time variable using the Fourier transform. Time variables using the Fourier transform will be able to effectively learn differences between times. As a result of evaluating the type of time variable in the summer electricity demand forecast for 2022 and 2023 using the BiGRU model, the model using the time variable using Fourier transform showed the best performance with MAPE of 2.01% and 2.04% confirmed. The results of this study are expected to improve prediction accuracy in the summer when power usage increases and prevent problems such as large-scale power outages.

높은 기온으로 인해 냉방부하가 상승하는 하절기의 낮에는 시간당 전력 수요가 증가하고 밤에는 낮에 비해 상대적으로 전력수요가 적은 특징을 보인다. 이러한 특징은 하계 전력 수요 예측에 있어 중요한 정보로 간주한다. 그러나 시간 정보를 단순히 더미변수로 표현할 경우, 모델은 시간의 변화를 학습하기보다는 단순히 시간대 간의 차이를 인식하는 데 그치게 된다. 본 연구에서는 이러한 문제를 해결하기 위해 푸리에 변환을 활용하여 시간 변수를 사용한 방식으로 접근하고자 한다. 푸리에 변환을 이용한 시간 변수는 시간 간의 차이를 효과적으로 학습할 수 있을 것이다. BiGRU 모델을 사용하여 2022년과 2023년의 하계 전력 수요예측에 시간 변수의 형식에 따른 평가를 수행한 결과 푸리에 변환을 사용한 시간 변수를 사용한 모델이 MAPE 2.01%, 2.04%로 가장 우수한 성능을 보이는 것을 확인하였다. 본 연구의 결과는 전력 사용량이 증가하는 하절기의 전력 사용량 예측 정확도를 높여, 대규모 정전 사태와 같은 문제를 방지할 수 있을 것으로 기대한다.

Keywords

Acknowledgement

This research was supported by the MSIT(Ministry of Science and ICT), Korea, under the ICAN(ICT Challenge and Advanced Network of HRD) support program(IITP-2024-RS-2022-00156409) supervised by the IITP(Institute for Information & Communications Technology Planning & Evaluation) and This work was supported by the Regional Innovation Strategy (RIS) program funded by the Ministry of Education in 2024 and managed by the National Research Foundation of Korea. (NRF Project Management Number: 2021RIS-003)

References

  1. J. D. Park, "Development of real-time demand forecasting techniques and research on online demand forecasting methods", https://new.kpx.or.kr/board.es?mid=a10502000000&bid=0045&act=view&list_no=51480&tag=&nPage=34 
  2. S. I. Kim, "Analysis of the impact of COVID-19 on electricity supply and demand in Korea", https://www.keei.re.kr/main.nsf/index.html?open&p=%2Fweb_keei%2Fd_results.nsf%2F0%2F89832dbd2f1f0c24492587a4002ddc3a&s=%3FOpenDocument%26is_popup%3D1%26menucode%3D1%26menucode%3DSS18%26Click%3D 
  3. H. W. Jung and K. B. Song, "Daily Maximum Electric Load Forecasting for the Next 4 Weeks for Power System Maintenance and Operation," The transactions of The Korean Institute of Electrical Engineers, Vol. 63, No. 11, pp. 1497-1502, Nov. 2014. DOI: 10.5370/KIEE.2014.63.11.1497 
  4. B. H. Ahn, H. R. Choi and H. C. Lee, "Regional Long-term/Mid-term Load Forecasting using SARIMA in South Korea," Journal of the Korea Academia-Industrial cooperation Society, Vol. 16, No. 12, pp. 8576-8584, Dec. 2015. DOI: 10.5762/KAIS.2015.16.12.8576 
  5. D. H. Shin and C. B. Kim, "A Study on Deep Learning Input Pattern for Summer Power Demand Prediction," The Journal of Korean Institute of Information Technology, Vol. 14, No. 11, pp. 127-134, 2016. DOI: 10.14801/jkiit.2016.14.11.127 
  6. H. B. Moon, J. K. Ahn and C. Y. Lee, "Forecasting electricity demand of Jeju island using machine learning," Innovation studies, Vol. 18, No. 4, pp. 49-63, Nov. 2023. DOI: 10.46251/INNOS.2023.11.18.4.49 
  7. P. S. Ji and J. Y. Lim, "Development of Daily Peak Power Demand Forecasting Algorithm Considering of Characteristics of Day of Week," THE TRANSACTION OF THE KOREAN INSTITUTE OF ELECTRICAL ENGINEERS P, Vol. 63P, No. 4, pp. 307-311, Dec. 2014, Doi: 10.5370/KIEEP.2014.63.4.307 
  8. J. H. Han and J. K. Baek, "The Load Forecasting in Summer Considering Day Factor," Journal of the Korea Academia-Industrial cooperation Society, Vol. 11, No. 8, pp. 2793-2800, Aug. 2010, DOI: 10.5762/KAIS.2010.11.8.2793 
  9. Y. S. Kim and H. J. Park, "Modeling and Predicting South Korea"s Daily Electric Demand Using DNN and LSTM," Journal of Climate Change Research, Vol. 12, No. 3, pp. 241-253, Jun. 2021. DOI: 10.15531/KSCCR.2021.12.3.241 
  10. Hoyong Eom, and Seung-Mook Baek, "Optimal Model Selection of Long Short-Term Memory for Mid-Term Electric Load Forecasting," The transactions of The Korean Institute of Electrical Engineers, Vol. 71, No. 11, pp. 1658-1665, Nov. 2022. DOI: 10.5370/KIEE.2022.71.11.1658 
  11. C. H. Moon, B. S. Kwon and K. B. Song, "A Deep Learning Model and Training Technique for 240 Hours Load Forecasting in Korea Power System," The transactions of The Korean Institute of Electrical Engineers, Vol. 71, No. 4, pp. 585-591, Apr. 2022. DOI: 10.5370/KIEE.2022.71.4.585 
  12. S. J. Ko, H. Y. Yoon and D. M. Shin, "Electronic Demand Data Prediction using Bidirectional Long Short Term Memory Networks," Journal of Software Assessment and Valuation, Vol. 14, No. 1, pp. 33-40, 2018. 
  13. S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," in Neural Computation, vol. 9, no. 8, pp. 1735-1780, 15 Nov. 1997, doi: 10.1162/neco.1997.9.8.1735. 
  14. K. Cho, B. V. Merrienboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, "Learning phrase representations using RNN Encoder-Decoder for statistical machine translation," arXiv:1406.1078, pp.1724-1734, Jun. 2014. 
  15. K. Cho, B. V. Merrienboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, "Learning phrase representations using RNN Encoder-Decoder for statistical machine translation," arXiv:1406.1078, pp.1724-1734, Jun. 2014. 
  16. H. U. Seol, "Comparison Study of LSTM and GRU Combination Models to Improve Summer Power Demand Prediction Performance" Graduate School of Gyeongsang National University, Feb. 2022, UCI: I804:48003-000000034410 
  17. EPSIS, https://epsis.kpx.or.kr 
  18. Y. R. Shin and S. H. Yoon, "Evaluation of weather information for electricity demand forecasting," Journal of the Korean Data And Information Science Society, Vol. 27, No. 6, pp. 1601-1607, Jul. 2016, DOI: https://doi.org/10.74 65/jkdi.2016.27.6.1601  https://doi.org/10.7465/jkdi.2016.27.6.1601
  19. Korea Meteorological Administration, https://data.kma.go.kr 
  20. B. C. Oh, H. J. Seo, H. K. Kank, S. Y. Kim, and M. A. Acquah, "Comparative Analysis of Artificial Intelligent Prediction Models for Nationwide Short-Term Electricity Demand," THE TRANSACTION OF THE KOREAN INSTITUTE OF ELECTRICAL ENGINEERS P, Vol. 69P, No. 4, pp. 253-259, 2020, DOI: 10.5370/KIEEP.253 
  21. S. h. Kwon. "A study of algorithms for predicting daily short-term electricity demand based on machine learning techniques." Graduate School of Wiedok University, Feb. 2021, UCI: I804:47032-200000508913