KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Development of a System for Predicting Photovoltaic Power Generation and Detecting Defects Using Machine Learning

기계학습을 이용한 태양광 발전량 예측 및 결함 검출 시스템 개발

  • 이승민 (경북대학교 컴퓨터학부) ;
  • 이우진 (경북대학교 컴퓨터학부/소프트웨어기술연구소)
  • Received : 2016.08.09
  • Accepted : 2016.08.29
  • Published : 2016.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, solar photovoltaic(PV) power generation which generates electrical power from solar panels composed of multiple solar cells, showed the most prominent growth in the renewable energy sector worldwide. However, in spite of increased demand and need for a photovoltaic power generation, it is difficult to early detect defects of solar panels and equipments due to wide and irregular distribution of power generation. In this paper, we choose an optimal machine learning algorithm for estimating the generation amount of solar power by considering several panel information and climate information and develop a defect detection system by using the chosen algorithm generation. Also we apply the algorithm to a domestic solar photovoltaic power plant as a case study.

여러 개의 태양전지들이 붙어 있는 태양광 패널을 이용하여 전력을 생산하는 태양광 발전은 최근 신재생 에너지 기술로 빠르게 성장하고 있는 분야이다. 하지만 태양광발전의 단점 중 하나인 불규칙한 전력 생산문제로 인해, 장비 및 패널 결함에 빠르게 대응하지 못하는 문제가 발생한다. 이 연구에서는 다양한 기후데이터와 패널 정보를 이용하여 태양광발전량 예측 방법들을 비교하여 최적의 예측 알고리즘을 평가하고 이를 기반으로 태양광발전소 결함 검출 시스템을 개발하여 국내 태양광 발전소에 적용한 사례를 기술한다.

Keywords

References

  1. Takefuji, Yoshiyasu, "Neural network parallel computing," Springer Science & Business Media, Vol.164, 2012.
  2. Harrington Peter, Machine learning in action, Manning, 2012.
  3. Jiquan Ngiam, "Multimodal deep learning," ICML-11 Conference on Machine Learning, pp.689-696, 2011.
  4. REN21, Renewables 2011 Global Status Report, REN21 Secretariat, 2011.
  5. Justine Sanchez, "Potential PV Problems & New Tools for Troubleshooting" [Internet], http://www.homepower.com/articles/solar-electricity/design-installation/potential-pv-problems/.
  6. Radial_basis_function_kernel [Internet], https://en.wikipedia.org/wiki/.
  7. N. Sharma, P. Sharma, D. Irwin, and P. Shenoy, "Predicting Solar Generation from Weather Forecasts Using Machine Learning," IEEE International Conference on Smart Grid Communications, 2011.
  8. Seung Min Lee and Woo Jin Lee, "Backpropagation Algorithm based Fault Detection Model of Solar Power Generation using Weather Data and Solar Power Generation Data," The 2015 Spring Conference of the KIPS, Vol.22, No.1, 2015(in Korean).
  9. Backpropagation [Internet], https://en.wikipedia.org/wiki/.
  10. Jae-Ju Song, Sang-Ho Lee, and Yoon-Su Jeong, "Analysis of prediction model for solar power generation," The Society of Digital Policy & Management, Vol.12, No.3, pp.243-248, 2014(in Korean).
  11. PVoutput [Internet], http://pvoutput.org/about.html/.
  12. Korea Meteorological Administration, Weather Resource Analysis Report for the Optimal Utilization of Solar Energy, 2008(in Korean).
  13. Changsong Chen, "Online 24-h solar power forecasting based on weather type classification using artificial neural network," Solar Energy, Vol.85, No.11, pp.2856-2870, 2011. https://doi.org/10.1016/j.solener.2011.08.027
  14. NOAA, Unedited Local Climatological Data Samples [Internet], http://www.ncdc.noaa.gov/ulcd/ULCD/.
  15. MesoWest [Internet], http://mesowest.utah.edu/.
  16. Tianfeng Chai and R. Roland Draxler, "Root mean square error (RMSE) or mean absolute error (MAE)?-Arguments against avoiding RMSE in the literature," Geoscientific Model Development, Vol.7, No.3, pp.1247-1250, 2014. https://doi.org/10.5194/gmd-7-1247-2014
  17. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: A simple way to prevent neural networks from overfitting," The Journal of Machine Learning Research, Vol.15, No.1, pp.1929-1958, 2014.
  18. A Library for Support Vector Machines [Internet], https://www.csie.ntu.edu.tw/-cjlin/libsvm/.
  19. Korea Meteorological Administration [Internet], http://www.kma.go.kr/.
  20. Java Neural Network Framework [Internet], http://neuroph.sourceforge.net/.