10.1007/s40095-022-00501-9

Predicting on-site solar energy generation using off-site weather stations and deep neural networks

  • Jose Ramirez-Vergara1,
  • L. B. Bosman*1,
  • Walter D. Leon-Salas1,
  • Ebisa Wollega2,
  1. Purdue Polytechnic Institute, Purdue University, West Lafayette, IN, 47907, US
  2. Department of Engineering, Colorado State University-Pueblo, Pueblo, CO, 81001, US

Published in Issue 2022-05-25

How to Cite

Ramirez-Vergara, J., Bosman, L. B., Leon-Salas, W. D., & Wollega, E. (2022). Predicting on-site solar energy generation using off-site weather stations and deep neural networks. International Journal of Energy and Environmental Engineering, 14(1 (March 2023). https://doi.org/10.1007/s40095-022-00501-9
Crossref
Scopus
Google Scholar
Europe PMC

Abstract

Abstract The growing trend of solar photovoltaic (PV) adoption has motivated homeowners and independent solar PV plants to assume the role of electricity generators. Utility companies need to address the changes in supply and demand to formulate effective distribution plans. Monitoring the power output of the photovoltaic array is a key task in ensuring the correct estimation of the electricity production of the system. Nevertheless, the literature recognizes the obstacle that expensive on-site monitoring sensors have on the cost of small-scale and medium-scale PV applications. Mathematical models based on the estimation of power output from historical data are the most used techniques to address the need for expensive monitoring systems. Estimating the power output of a PV array depends on the cell temperature, ambient temperature, and solar irradiance. This paper proposes a machine learning model to forecast site-specific ambient temperature and solar irradiance. The results contribute to the generation of low-cost data-driven models to save money by eliminating the need to install on-site sensors. The methodology employed off-site publicly-available weather data from neighboring weather stations as an alternative to on-site measurements. A 5-layer deep neural network was trained using 5 years’ worth of historical data from a remote weather station in Green Bay, WI, to predict on-site parameters for a solar array located about 45 miles away in Keshena, WI. The model was found to be suitable for site-specific ambient temperature prediction. The ozone Dobson was a key parameter for solar irradiance prediction, given that the predictive accuracy of the proposed model was limited by the size of the training data.

Keywords

  • Deep neural networks,
  • Weather stations,
  • Photovoltaic,
  • Prediction,
  • Monitoring

References

  1. Abreu et al. (2019) New trends in solar: a comparative study assessing the attitudes towards the adoption of rooftop PV (pp. 347-363) https://doi.org/10.1016/j.enpol.2018.12.038
  2. Lukanov and Krieger (2019) Distributed solar and environmental justice: exploring the demographic and socio-economic trends of residential PV adoption in California https://doi.org/10.1016/j.enpol.2019.110935
  3. Ramirez, J., Soto, E., Wollega, E., Bosman, L.: Using machine learning to assess solar energy grid disturbances. In: Proceedings of the International Conference on Industrial Engineering and Operations Management
  4. .
  5. Detroit, Michigan, USA, August 9–11, (2020)
  6. Bosman et al. (2020) PV System Predictive maintenance: challenges, current approaches, and opportunities https://doi.org/10.3390/en13061398
  7. Ayvazogluyuksel and Filik (2018) Estimation methods of global solar radiation, cell temperature and solar power forecasting: a review and case study in Eskisehir (pp. 639-653) https://doi.org/10.1016/j.rser.2018.03.084
  8. Pazikadin, A.R., Rifai, D., Ali, K., Malik, M.Z., Abdalla, A.N., Faraj, M.A.: Solar irradiance measurement instrumentation and power solar generation forecasting based on Artificial Neural Networks (ANN): A review of five years research trend. The Science of the total environment
  9. ,
  10. vol. 715, pp. 136848–136848, Jan-22 (2020)
  11. Meleti et al. (2009) Factors affecting solar ultraviolet irradiance measured since 1990 at Thessaloniki, Greece (pp. 4167-4179) https://doi.org/10.1080/01431160902822864
  12. Ghiani et al. (2013) Evaluation of photovoltaic installations performances in Sardinia (pp. 1134-1142) https://doi.org/10.1016/j.enconman.2013.09.012
  13. Kalogirou (2001) Artificial neural networks in renewable energy systems applications: a review (pp. 373-401) https://doi.org/10.1016/S1364-0321(01)00006-5
  14. Obando et al. (2019) Solar radiation prediction using machine learning techniques: a review (pp. 684-697) https://doi.org/10.1109/TLA.2019.8891934
  15. Ju et al. (2013) An improved temperature estimation method for solar cells operating at high concentrations (pp. 80-89) https://doi.org/10.1016/j.solener.2013.02.028
  16. TamizhMani, G., Ji, L., Tang, Y., Petacci, L., Osterwald, C.: Photovoltaic Module Thermal/Wind Performance: Long-term Monitoring and Model Development for Energy Rating, presented at the NCPV and Solar Program Review Meeting 2003, (2003)
  17. Law et al. (2014) Direct normal irradiance forecasting and its application to concentrated solar thermal output forecasting—a review (pp. 287-307) https://doi.org/10.1016/j.solener.2014.07.008
  18. Krishnamurti, T.N.: Numerical weather prediction. In: Annual Review of Fluid Mechanics. vol. 27, Lumley, J.L. Van Dyke, M. (eds.) ed: Annual Reviews Inc. {a}, P.O. Box 10139, 4139 El Camino Way, Palo Alto, California 94306, USA, pp. 195–224 (1995)
  19. Bouabbou, A., Ghennioui, A., Vaudreuil, S., Naimi, Z.: Short-term solar irradiance prediction using Time series analysis and Neural Networks for Green Energy Park Photovoltaic Plant. In: Proceedings of the 11th Ises Eurosun 2016 Conference
  20. ,
  21. pp. 1447–1458 (2017)
  22. Box et al. (2015) John Wiley & Sons
  23. Zhang (2003) Time series forecasting using a hybrid ARIMA and neural network model (pp. 159-175) https://doi.org/10.1016/S0925-2312(01)00702-0
  24. Brabec et al. (2015) Tailored vs black-box models for forecasting hourly average solar irradiance (pp. 320-331) https://doi.org/10.1016/j.solener.2014.11.003
  25. Zeng and Qiao (2013) Short-term solar power prediction using a support vector machine (pp. 118-127) https://doi.org/10.1016/j.renene.2012.10.009
  26. Sharma et al. (2016) Short term solar irradiance forecasting using a mixed wavelet neural network (pp. 481-492) https://doi.org/10.1016/j.renene.2016.01.020
  27. Alanazi, M., Khodaei, A.: Day-ahead solar forecasting using time series stationarization and feed-forward neural network. In: Gao, D.W., Muljadi, E., Zhang, J., Khodaei, A. (eds.) 2016 North American Power Symposium (2016)
  28. Alzahrani, A., Shamsi, P., Dagli, C., Ferdowsi, M.: Solar irradiance forecasting using deep neural networks. In: Complex Adaptive Systems Conference with Theme: Engineering Cyber Physical Systems
  29. ,
  30. Cas
  31. ,
  32. vol. 114, pp. 304–313 (2017)
  33. Capizzi et al. (2012) Innovative second-generation wavelets construction with recurrent neural networks for solar radiation forecasting (pp. 1805-1815) https://doi.org/10.1109/TNNLS.2012.2216546
  34. Dehini and Berbaoui (2018) Solar energy control and power quality improvement using multilayer feed forward neural network (pp. 1954-1962)
  35. Mellit and Pavan (2010) A 24-h forecast of solar irradiance using artificial neural network: application for performance prediction of a grid-connected PV plant at Trieste, Italy (pp. 807-821) https://doi.org/10.1016/j.solener.2010.02.006
  36. Basheer and Hajmeer (2000) Artificial neural networks: fundamentals, computing, design, and application (pp. 3-31) https://doi.org/10.1016/S0167-7012(00)00201-3
  37. Cao and Cao (2005) Forecast of solar irradiance using recurrent neural networks combined with wavelet analysis (pp. 161-172) https://doi.org/10.1016/j.applthermaleng.2004.06.017
  38. Gutierrez-Corea et al. (2016) Forecasting short-term solar irradiance based on artificial neural networks and data from neighboring meteorological stations (pp. 119-131) https://doi.org/10.1016/j.solener.2016.04.020
  39. Husein and Chung (2019) Day-Ahead solar irradiance forecasting for microgrids using a long short-term memory recurrent neural network: a deep learning approach https://doi.org/10.3390/en12101856
  40. Fan et al. (2018) Comparison of support vector machine and extreme gradient Boosting for predicting daily global solar radiation using temperature and precipitation in humid subtropical climates: a case study in China (pp. 102-111) https://doi.org/10.1016/j.enconman.2018.02.087
  41. Paredes-Parra et al. (2018) PV module monitoring system based on low-cost solutions: wireless Raspberry application and assessment https://doi.org/10.3390/en11113051
  42. World Weather Online. (2021, 02/17).
  43. World Weather
  44. . Available:
  45. https://www.worldweatheronline.com/
  46. Sperling's Best Places. (n.d., 03/25).
  47. Climate in Keshena, Wisconsin
  48. . Available:
  49. https://www.bestplaces.net/climate/city/wisconsin/keshena
  50. Weather Spark. (2021, 03/25).
  51. Average Weather in Keshena
  52. . Available:
  53. https://weatherspark.com/y/13587/Average-Weather-in-Keshena-Wisconsin-United-States-Year-Round#:~:text=In%20Keshena%2C%20the%20summers%20are,or%20above%2090%C2%B0F
  54. Sperling's Best Places. (n.d., 03/25).
  55. Weather in Green Bay, Wisconsin
  56. . Available:
  57. https://www.bestplaces.net/climate/city/wisconsin/green_bay
  58. Weather Spark. (2021, 03/25).
  59. Average Weather in Green Bay
  60. . Available:
  61. https://weatherspark.com/y/13582/Average-Weather-in-Green-Bay-Wisconsin-United-States-Year-Round
  62. Apple Inc. (2021, 05/31).
  63. Dark Sky
  64. . Available:
  65. https://darksky.net/
  66. Brogden (1946) On the interpretation of the correlation coefficient as a measure of predictive efficiency https://doi.org/10.1037/h0061548
  67. Hossen, T., Plathottam, S.J., Angamuthu, R.K., Ranganathan, P., Salehfar, H.: Short-term load forecasting using deep neural networks (DNN). In 2017 North American Power Symposium (NAPS), pp. 1--6 (2017)
  68. Shi et al. (2017) Deep learning for household load forecasting—a novel pooling deep RNN (pp. 5271-5280) https://doi.org/10.1109/TSG.2017.2686012
  69. Merkel, G.: Deep Neural Networks as Time Series Forecasters of Energy Demand (2017)
  70. Liu et al. (2017) A survey of deep neural network architectures and their applications (pp. 11-26) https://doi.org/10.1016/j.neucom.2016.12.038
  71. University of New South Wales. (n.d., 03/21). Backpropagation. Available:
  72. https://www.cse.unsw.edu.au/~cs9417ml/MLP2/BackPropagation.html
  73. Facebook Inc. (2020, 03/21). PyTorch. Available:
  74. https://pytorch.org/
  75. Chai and Draxler (2014) Root mean square error (RMSE) or mean absolute error (MAE) (pp. 1525-1534)
  76. Willmott and Matsuura (2005) Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance (pp. 79-82) https://doi.org/10.3354/cr030079
  77. Barrett (1974) The coefficient of determination—some limitations (pp. 19-20)
  78. Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z.B., Swami, A.: The limitations of deep learning in adversarial settings. In: 2016 IEEE European Symposium on Security and Privacy (EuroS&P), pp. 372–387 (2016)