All Issue

2019 Vol.56, Issue 2 Preview Page

Research Paper

Estimation of Solar Irradiance at Weather Stations in Korea Using Regionally Trained Artificial Neural Network Models

권역별 인공신경망 모델을 이용한 전국 기상관측소의 일사량 추정

YoungHyun Koo1
,
Sung-Min Kim2
,
Myeongchan Oh1
,
Hyeong-Dong Park13*
구 영현1
,
김 성민2
,
오 명찬1
,
박 형동13*
1서울대학교 공과대학 에너지시스템공학부
2강원대학교 공과대학 에너지공학부(에너지자원융합공학전공)
3서울대학교 에너지자원신기술연구소
* Corresponding Author
30 April 2019. pp. 155-171
PDF XML Citation
Abstract

The ability to estimate solar irradiance is necessary for efficient installation and management of solar photovoltaic systems. Weather observation data and Communication, Ocean and Meteorological Satellite (COMS) images were used to estimate solar irradiance at 94 Korea Meteorological Administration (KMA) weather stations. First, Korea was divided into 13 regions by applying K-means clustering to cloud indices derived from visible channel images of COMS, and artificial neural network (ANN) models were trained in 10 regions. Ten variables were selected as inputs for ANN by mutual information feature selection to build multi-layer perceptron networks in MATLAB. The cross validation resulted in a 0.96 correlation coefficient and 80.87 W/m2 (22.62%) root mean square error (RMSE). The regionally trained ANN models could reduce the relative RMSE by 2%p compared to non-regionally trained models. The solar irradiances at the 94 KMA weather stations were estimated by applying these models. The solar irradiance of points that have no measured data from a pyranometer can be estimated using these regional ANN models.

Keywords
Solar irradiance
Artificial Neural Network
Weather station
Satellite images
Region classification

일사량을 파악하는 것은 태양광 발전 시스템을 설치하고 운영하는 데 있어서 필수적이다. 본 연구에서는 기상청에서 운영하고 있는 94개 기상관측소의 일사량을 추정하기 위해 기상 관측 자료와 천리안 위성영상을 이용하였다. 우선, 천리안 위성의 가시광 채널 영상으로부터 산출한 구름지수에 K-평균 군집화를 적용하여 한국을 13개의 권역으로 분류하였으며, 이 중 일사량 관측값이 존재하는 10개 권역의 일사량 추정을 위해 인공신경망 모델을 훈련시켰다. 인공신경망의 입력변수로는 상호정보량 특징선택을 통해 총 10개의 변수가 선택되었고, MATLAB 소프트웨어를 활용하여 인공신경망 모델을 구축하였다. 교차검증 결과, 본 연구의 모델은 상관계수 0.96, RMSE 80.87 W/m2 (22.62%)의 오차를 보였으며, 권역 분류를 적용하지 않은 전국적 훈련모델에 비해 오차가 약 2%p 감소하는 효과가 있었다. 이러한 권역별 일사량 모델을 활용하여 기상청의 기상관측소 94개 지점의 일사량을 추정하였다. 본 연구의 권역별 인공신경망 모델을 활용하면 일사계는 없지만 기본적인 기상요소가 관측되고 있는 관측소의 실시간 일사량을 20% 초반대의 오차 수준으로 추정할 수 있을 것으로 기대된다.

키워드
일사량
인공신경망
기상관측소
위성영상
권역 분류
References
1
Assouline, D., Mohajeri, N., and Scartezzini, J.L., 2017. Quantifying rooftop photovoltaic solar energy potential: A machinelearning approach. Solar Energy, 141, 278-296.
10.1016/j.solener.2016.11.045
2
Battiti, R., 1994. Using mutual information for selecting features in supervised neural net learning.  IEEE Transactions on Neural Networks, 5(4), 537-550.
10.1109/72.29822418267827
3
Bechini, L., Ducco, G., Donatelli, M., and Stein, A., 2000. Modelling, interpolation and stochastic simulation in space and time of global solar radiation, Argriculture. Ecosystem and Environment, 8, 29-42.
10.1016/S0167-8809(00)00170-5
4
Bezzi, M. and Vitti, A. 2005. A comprison of some kriging inter-polation methods for the production of solar radiation maps, Geomatics Workbooks, 1-17.
5
Bishop, C.M., 1995. Neural Networks for Pattern Recognition, Oxford University Press, New York, 482p.
6
Calinski, T. and Harabasz, J., 1974. A dendrite method for clusteranalysis. Communications in Statistics, 3, 1-27.
10.1080/03610917408548446
7
Choi, W.S., Song, A.R., and Kim, Y.I., 2015. Solar irradiance estimation in Korea by using modified Heliosat-II method and COMS-MI imagery. J. the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 33(5), 463-472.
10.7848/ksgpc.2015.33.5.463
8
Dorvlo, A.S.S., Jervase, J.A., and Al-Lawati, A., 2002. Solar radiation estimatino using artificial neural networks. Applied Energy, 71, 307-319.
10.1016/S0306-2619(02)00016-8
9
Eissa, Y., Marpu, P.R., Gherboudj, I., Ghedira, H., Ouarda, T.B.M.J., and Chiesa, M., 2013. Artificial neural network based model for retrieval of the direct normal, diffuse horizontal and global horizontal irradiances using SEVIRI images. Solar Energy, 89, 1-16.
10.1016/j.solener.2012.12.008
10
Gardner, M.W. and Dorling, S.R., 1998. Artificial neural networks (the multilayer perceptron) - a review of applications in the atmosphere sciences. Atmospheric Environment, 32(14), 2627-2636.
10.1016/S1352-2310(97)00447-0
11
Hammer, A., Heinemann, D., Hoyer, C., Kuhlemann, R., Lorenz, E., Muller, R., and Beyer, H.G., 2003. Solar energy assess-ment using remote sensing technologies. Remote Sensing of Environment, 86, 423-432.
10.1016/S0034-4257(03)00083-X
12
Haykin, S.O., 2009. Neural Network and Learning Machines, Pearson, 936p.
13
Jamaly, M. and Kleissl, J., 2017. Spatiotemporal interpolation and forecast of irradiance data using Kriging. Solar Energy, 158, 407-423.
10.1016/j.solener.2017.09.057
14
Jolliffe, I.T., 1986. Principal Component Analysis, Springer, 487p.
10.1007/978-1-4757-1904-8PMC1960413
15
Journee, M. and Bertrand, C., 2010. Improving the spatio-temporal distribution of surface solar radiation data by merging ground and satellite measurements. Remote Sensing of Environment, 114, 2692-2704.
10.1016/j.rse.2010.06.010
16
Kaastra, L. and Boyd, M., 1996. Designing a neural network for forecasting financial and economic time series. Neurocomputing, 10, 215-236.
10.1016/0925-2312(95)00039-9
17
Kim, C.K., Kim, H., Kang, Y. and Yun, C., 2017. Toward improved solar irradiance forecasts: comparison of the global horizontal irradiances derived from the COMS satellite imagery over the Korean peninsula. Pure Applied Geophysics, 174, 2773-2792.
10.1007/s00024-017-1578-y
18
Kim, H. and Kim, J., 2016. Correlation to predict global solar insolation and evaluation of that correlation for Korea(I). New & Renewable Energy, 12(S2), 30-35.
10.7849/ksnre.2016.10.12.S2.30
19
Kim, J., Suh, J., Park, H., Kang, Y., and Yun, C., 2014. Assessment of photovoltaic potential for constructing Ulleung-do as a sustainable energy island. J. Korean Society of Mineral and Energy Resources Engineering, 51(4), 513-524.
10.12972/ksmer.2014.51.4.513
20
Lee, J., Choi, W., Kim, Y., Yun, C., Jo, D., and Kang, Y., 2013. Estimation of global horizontal insolation over the Korean peninsula based on COMS MI satellite images. Korean J. Remote Sensing, 29(1), 151-160.
10.7780/kjrs.2013.29.1.14
21
Linares-Rodriguez, A., Ruiz-Arias, J.A., Pozo-Vaquez, D., and Tovar-Pescador, J., 2013. An artificial neural network ensemble model for estimating global solar radiation from Meteosat satellite images. Energy, 61, 636-645.
10.1016/j.energy.2013.09.008
22
Liou, K.N., 2002. An Introductin to Atmospheric Radiation (2nd Edition), Academic Press, San Diego, USA, 583p.
23
MacKay, D.J.C., 1992. Bayesian interpolation. Neural computation, 4(3), 415-447.
10.1162/neco.1992.4.3.415
24
MacQueen, J.B., 1967. Some Methods for classification and analysis of multivariate observations. Proceeding of 5thBerkeley Symposium on Mathematical Statistics and Probability, 1, 281-297.
25
May R., Dandy, G., and Maier, H., 2011. Review of Input Variables Selection Methods for Artificial Neural Networks In Artificial Neural Networks - Methodological Advance and Biomedical Applications, IntechOpen, London, UK, 19-44.
10.5772/16004
26
Noia, M, Ratto, C.F., and Festa, R., 1993a. Solar irradiance estimation from geostationary satellite data: Statistical models. Solar Energy, 51(6), 449-456.
10.1016/0038-092X(93)90130-G
27
Noia, M, Ratto, C.F., and Festa, R., 1993b. Solar irradiance estimation from geostationary satellite data: Physical models. Solar Energy, 51(6), 457-465.
10.1016/0038-092X(93)90131-7
28
Palmer, D., Cole, I., Betts, T., and Gottschalg, R., 2017. Interpolating and estimating horizontal diffuse solar irradiation  to provide UK-Wide coverage: selection of the best performing models. Energies, 10(181), 1-23.
10.3390/en10020181
29
Peng H., Long, F., and Ding, C., 2005. Feture selection based on mutual information: criteria ofmax-dependency, max-relevance, and min-redundancy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(8), 1226-1238.
10.1109/TPAMI.2005.15916119262
30
Perez, R., Ineichen, P., Moore, K., Kmiecik, M., Chain, C., George, R., and Vignola, F., 2002. A new operational model for satellite-derived irradiances: description and validation. Solar Energy, 73(5), 307-317.
10.1016/S0038-092X(02)00122-6
31
Rigollier, C., Bauer, O., and Wald, L., 2000. On the clear sky model of the ESRA - European solar radiation atlas -with respect to the heliosat method. Solar Energy, 68(1), 33-48.
10.1016/S0038-092X(99)00055-9
32
Rigollier, C., Lefevre, M., and Wald, L., 2004. The method Heliosat-2 for deriving shortwave solar radiation from satellite images. Solar Energy, 77, 159-169.
10.1016/j.solener.2004.04.017
33
Salvador, S. and Chan, P., 2005. Learning states and rules for detecting anomalies in time series. Applied Intelligence, 23(3), 2170-2176.
10.1007/s10489-005-4610-3
34
Sen, Z. and Sahin, A.D., 2001. Spatial interpolation and estimation of solar irradiation by cumulative semivariograms. Solar Energy, 71(1), 11-21.
10.1016/S0038-092X(01)00009-3
35
Shannon, C.E., 1948. A mathematical theory of communication. Bell System Technological J., 27, 379-423.
10.1002/j.1538-7305.1948.tb01338.x
36
Song, J. and Choi, Y., 2012. Assessment of rooftop photovoltaic potential in the Pukyong National University for establishing a green campus. J. Korean Society of Mineral and Energy Resources Engineering, 49(3), 369-378.
37
Song, J., Choi, Y., Cheong, Y., Yim, G., and Park, H., 2013. Site suitability assessment for installing photovoltaic systems using fisheye-lens camera and SOLPOS calculator. J. Korean Society of Mineral and Energy Resources Engineering, 50(6), 789-798.
10.12972/ksmer.2013.50.6.789
38
Song, J., Choi, Y., Jang, M., and Yoon, S., 2014. A comparison of wind power and photovoltaic potentials at Yeongok, Mulno and Booyoung abandoned mines in Kangwon province, Korea. J. Korean Society of Mineral and Energy Resources Engineering, 51(4), 525-536.
10.12972/ksmer.2014.51.4.525
39
Tapakis, R., and Charalambides, A.G., 2014. Enhanced values of global irradiance due to the presence of clouds in Eastern Mediterranean. Renewable Energy, 62, 459-467.
10.1016/j.renene.2013.08.001
40
Verbois, H., Huva, R., Rusydi, A., and Walsh, W., 2018. Solar irradiance forecasting in the tropics using numerical weather prediction and statistical learning. Solar Energy, 162, 265-277.
10.1016/j.solener.2018.01.007
41
Yang, D., Gu, C., Dong, Z., Jirutitijaroen, P., Chen, N., and Walsh, W.M., 2013. Solar irradiance forecasting using spatial-temporal covariance structures and time-forward kriging. Renewable Energy, 60, 235-245.
10.1016/j.renene.2013.05.030
42
Zagouras, A., Inman, R.H., Coimbra, C.F.M., 2014. On the determination of coherent solar microclimates for utility planning and operations. Solar Energy, 102, 173-188.
10.1016/j.solener.2014.01.021
43
Zagouras, A., Kazantzidis, A., Nikitidou, E., Argiriou, A.A., 2013. Determination of measuring sites for solar irradiance, based on cluster analysis of satellite-derived cloud estimations. Solar Energy, 97, 1-11.
10.1016/j.solener.2013.08.005
Information
  • Publisher :The Korean Society of Mineral and Energy Resources Engineers
  • Publisher(Ko) :한국자원공학회
  • Journal Title :Journal of the Korean Society of Mineral and Energy Resources Engineers
  • Journal Title(Ko) :한국자원공학회지
  • Volume : 56
  • No :2
  • Pages :155-171
  • Received Date : 2019-01-06
  • Revised Date : 2019-02-16
  • Accepted Date : 2019-04-29
  • DOI :https://doi.org/10.32390/ksmer.2019.56.2.155
Journal Informaiton Journal of the Korean Society of Mineral and Energy Resources Engineers Journal of the Korean Society of Mineral and Energy Resources Engineers
Online Submission
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • KCI 문헌 유사도 검사
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close