Abstract
The energy yield estimation of a photovoltaic (PV) system operating under partially shaded conditions is a challenging task and a very active area of research. In this paper, we attack this problem with the aid of machine learning techniques. Using data simulated by the equivalent circuit of a PV string operating under partial shading, we train and evaluate three different gradient boosted regression tree models to predict the global maximum power point (MPP). Our results show that all three approaches improve upon the state-of-the-art closed-form estimates, in terms of both average and worst-case performance. Moreover, we show that even a small number of training examples is sufficient to achieve improved global MPP estimation. The methods proposed are fast to train and deploy and allow for further improvements in performance should more computational resources be available.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Standard Test Conditions: 1000 W/m\(^2\) irradiance, 25 \(^{\circ }\)C temperature & 1.5 air mass.
- 2.
Initial attempts at exploiting interactions between \(P_{max}\) & \(V_{P_{max}}\) by first predicting the value of one and then using it to predict the other, yielded worse results than assuming independence. Their further investigation is left for future work.
- 3.
As we will see in this section, \(\hat{X}\) can be an estimate of \(P_{max}\) or \(V_{P_{max}}\), or of any of the ‘intermediate outputs’ –\(P_1\), \(V_1\), \(P_2\), \(V_2\)– depending on the method.
- 4.
Denoting MPP1 & MPP2 with ‘1’ & ‘\(-1\)’, respectively, the classifier’s prediction is of the form \(\hat{H} = sign\big [\sum _{m=1}^M {a_m h_m(G, T, s, n_{sh})}\big ]\in \{-1,1\}\), where \(h_m(G, T, s, n_{sh})\in \{-1,1\}\) is the prediction of the base learner added on the m-th round and \(a_m\) its voting weight, both the learner and \(a_m\) being the learned parameters of the model.
- 5.
- 6.
More precisely, \(95\%\) of the times, \(99.99\%\) of the estimates of \(P_{max}\) under the Stagewise model will have a NAE smaller than some value that lies between \(5.03\%\) and \(5.93\%\). In the discussion, we sacrifice this level of mathematical rigour for simplicity.
References
Batzelis, E., Kampitsis, G.E., Papathanassiou, S.A.: A MPPT algorithm for partial shading conditions employing curve fitting. In: EU PVSEC, pp. 1502–1507 (2016)
Batzelis, E.I., Georgilakis, P.S., Papathanassiou, S.A.: Energy models for photovoltaic systems under partial shading conditions: a comprehensive review. IET Renew. Power Gener. 9(4), 340–349 (2015)
Batzelis, E.I., Routsolias, I.A., Papathanassiou, S.A.: An explicit PV string model based on the Lambert W function and simplified MPP expressions for operation under partial shading. IEEE Trans. Sustain. Energy 5(1), 301–312 (2014)
Brecl, K., Topič, K., Topič, M.: Self-shading losses of fixed free-standing PV arrays. Renew. Energy 36(11), 3211–3216 (2011)
Busa-Fekete, R., Kégl, B., Éltető, T., Szarvas, G.: Ranking by calibrated AdaBoost. In: Proceedings of the Learning to Rank Challenge, pp. 37–48 (2011)
Caruana, R., Niculescu-Mizil, A.: An empirical comparison of supervised learning algorithms. In: ICML, pp. 161–168 (2006)
Chen, T., Guestrin, C.: Xgboost: a scalable tree boosting system. In: SIGKDD, pp. 785–794 (2016)
Deline, C., Dobos, A., Janzou, S., Meydbray, J., Donovan, M.: A simplified model of uniform shading in large photovoltaic arrays. Sol. Energy 96, 274–282 (2013)
Dolan, J.A., Lee, E., Yeh, C., Ben-Menahem, S., Ishihara, A.K.: Neural network estimation of photovoltaic IV curves under partially shaded conditions. In: IJCNN, pp. 1358–1365 (2011)
Fernández-Delgado, M., Cernadas, E., Barro, S., Amorim, D.: Do we need hundreds of classifiers to solve real world classification problems? JMLR 15, 3133–3181 (2014)
Friedman, J.H.: Greedy function approximation: a gradient boosting machine. Ann. Stat. 29, 1189–1232 (2000)
Mason, L., Baxter, J., Bartlett, P., Frean, M.: Boosting algorithms as gradient descent. In: NIPS, pp. 512–518. MIT Press (2000)
Moballegh, S., Jiang, J.: Modeling, prediction, and experimental validations of power peaks of PV arrays under partial shading conditions. Sustain. Energy 5(1), 293–300 (2014)
Nguyen, D.D., Lehman, B., Kamarthi, S.: Performance evaluation of solar photovoltaic arrays including shadow effects using neural network. IEEE Energy Convers. Congr. Expo. 6(2), 3357–3362 (2009)
Nikolaou, N., Edakunni, N., Kull, M., Flach, P., Brown, G.: Cost-sensitive boosting algorithms: do we really need them? Mach. Learn. 104(2), 359–384 (2016)
Psarros, G., Batzelis, E., Papathanassiou, S.: Analysis of local MPPs on the P-V curve of a partially shaded PV string. In: EU PVSEC, pp. 3383–3389 (2014)
Psarros, G.N., Batzelis, E.I., Papathanassiou, S.A.: Partial shading analysis of multistring PV arrays and derivation of simplified MPP expressions. IEEE Trans. Sustain. Energy 6(2), 499–508 (2015)
Rodrigo, P., Fernández, F., Almonacid, F., Pérez-Higueras, J.: A simple accurate model for the calculation of shading power losses in photovoltaic generators. Sol. Energy 93, 322–333 (2013)
Viola, P., Jones, M.J.: Robust real-time face detection. IJCV 57(2), 137–154 (2004)
Wyner, A.J., Olson, M., Bleich, J., Mease, D.: Explaining the success of adaboost and random forests as interpolating classifiers (2017). arXiv:1504.07676v2
Acknowledgements
This project was partially supported by the EPSRC Centre for Doctoral Training [EP/I028099/1] & the EPSRC LAMBDA [EP/N035127/1] & Anyscale Apps [EP/L000725/1] project grants. N. Nikolaou acknowledges the support of the EPSRC Doctoral Prize Fellowship. E. Batzelis carried out this research at NTUA, Athens, Greece under the support of the ‘IKY Fellowships of Excellence for Postgraduate Studies in Greece-Siemens Program’.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Nikolaou, N., Batzelis, E., Brown, G. (2017). Gradient Boosting Models for Photovoltaic Power Estimation Under Partial Shading Conditions. In: Woon, W., Aung, Z., Kramer, O., Madnick, S. (eds) Data Analytics for Renewable Energy Integration: Informing the Generation and Distribution of Renewable Energy. DARE 2017. Lecture Notes in Computer Science(), vol 10691. Springer, Cham. https://doi.org/10.1007/978-3-319-71643-5_2
Download citation
DOI: https://doi.org/10.1007/978-3-319-71643-5_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-71642-8
Online ISBN: 978-3-319-71643-5
eBook Packages: Computer ScienceComputer Science (R0)