Smart Storage Scheduling and Forecasting for Peak Reduction on Low-Voltage Feeders

  • Timur Yunusov
  • Georgios Giasemidis
  • Stephen Haben
Chapter
Part of the Studies in Systems, Decision and Control book series (SSDC, volume 149)

Abstract

The transition to a low carbon economy will likely bring new challenges to the distribution networks, which could face increased demands due to low-carbon technologies and new behavioural trends. A traditional solution to increased demand is network reinforcement through asset replacement, but this could be costly and disruptive. Smart algorithms combined with modern technologies can lead to inexpensive alternatives. In particular, battery storage devices with smart control algorithms can assist in load peak reduction. The control algorithms aim to schedule the battery to charge at times of low demand and discharge, feeding the network, at times of high load. This study analyses two scheduling algorithms, model predictive control (MPC) and fixed day-ahead scheduler (FDS), comparing against a set-point control (SPC) benchmark. The forecasts presented here cover a wide range of techniques, from traditional linear regression forecasts to machine learning methods. The results demonstrate that the forecasting and control methods need to be selected for each feeder taking into account the demand characteristics, whilst MPC tends to outperform the FDS on feeders with higher daily demand. This chapter contributes in two main directions: (i) several forecasting methods are considered and compared and (ii) new energy storage control algorithm, MPC with half-hourly updated (rolling) forecasts designed for low voltage network application, is introduced, analysed and compared.

Notes

Acknowledgements

The research work presented in this chapter have been initiated as a part of New Thames Valley Vision (SSET203), a Low Carbon Network Fund project, funded by Ofgem and led by Scottish and Southern Electricity Networks.

References

  1. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.  https://doi.org/10.1023/a:1010933404324.
  2. Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (1984). Classification and regression trees. Statistics/probability series. Belmont, California, USA: Wadsworth Publishing Company.Google Scholar
  3. Drucker, H., Burges, C. J. C., Kaufman, L., Smola, A. J., & Vapnik, V. (1997). Support vector regression machines. In M. C. Mozer, M. I. Jordan, & T. Petsche (Eds.) Advances in neural information processing systems (Vol. 9, pp. 155–161). MIT Press. URL http://papers.nips.cc/paper/1238support-vector-regression-machines.pdf.
  4. Dudek, G. (2015). Short-term load forecasting using Random forests. In Intelligent systems’ 2014. Advances in intelligent systems and computing (Vol. 323, pp. 821–828).Google Scholar
  5. Evans, G. (2016). New thames valley vision technical impact evaluation impact on DNO network from low carbon promotions, SDRC 9.8b (SSET203). In Technical Report 3, Scottish and Southern Energy Power Distribution.Google Scholar
  6. Gajowniczek, K., & Zbkowski, T. (2017). Electricity forecasting on the individual household level enhanced based on activity patterns. PLOS ONE, 12(4), 1–26.  https://doi.org/10.1371/journal.pone.0174098.
  7. García, C. E., Prett, D. M., & Morari, M. (1989). Model predictive control: Theory and practice-A survey. Automatica, 25(3), 335–348.  https://doi.org/10.1016/0005-1098(89)90002-2.CrossRefMATHGoogle Scholar
  8. Giasemidis, G., Haben, S., Lee, T., Singleton, C., & Grindrod, P. (2017). A genetic algorithm approach for modelling low voltage network demands. Applied Energy, 203, 463–473.  https://doi.org/10.1016/j.apenergy.2017.06.057.CrossRefGoogle Scholar
  9. Haben, S., & Giasemidis, G. (2016). A hybrid model of kernel density estimation and quantile regression for gefcom2014 probabilistic load forecasting. International Journal of Forecasting, 32, 1017–1022.CrossRefGoogle Scholar
  10. Hida, Y., Yokoyama, R., Shimizukawa, J., Iba, K., Tanaka, K., & Seki, T. (2010). Load following operation of NAS battery by setting statistic margins to avoid risks. IEEE PES General Meeting, PES, 2010, 1–5.  https://doi.org/10.1109/PES.2010.5588170.Google Scholar
  11. Hong, W. C. (2009). Electric load forecasting by support vector model. Applied Mathematical Modelling 33(5), 2444–2454.  https://doi.org/10.1016/j.apm.2008.07.010. URL http://www.sciencedirect.com/science/article/pii/S0307904X08001844.
  12. Hu, Z., Bao, Y., & Xiong, T. (2013). Electricity load forecasting using support vector regression with memetic algorithms. The Scientific World Journal, 2013.Google Scholar
  13. Joshi, K. A., & Pindoriya, N. M. (2015). Day-ahead dispatch of battery energy storage system for peak load shaving and load leveling in low voltage unbalance distribution networks. In Power & Energy Society General Meeting, 2015 IEEE (pp. 1–5).  https://doi.org/10.1109/pesgm.2015.7285673.
  14. Lahouar, A., & Slama, J. B. H. (2015). Random forests model for one day ahead load forecasting. In Renewable Energy Congress (IREC), 2015 6th International. IEEE, Sousse, Tunisia.Google Scholar
  15. Megel, O., Mathieu, J. L., & Andersson, G. (2015). Scheduling distributed energy storage units´ to provide multiple services under forecast error. International Journal of Electrical Power & Energy Systems, 72, 48–57.  https://doi.org/10.1016/j.ijepes.2015.02.010. URL http://linkinghub.elsevier.com/retrieve/pii/S0142061515000939.
  16. Nair, N., & Garimella, N. (2010). Battery energy storage systems: Assessment for small-scale renewable energy integration. Energy and Buildings, 42, 2124–2130.CrossRefGoogle Scholar
  17. Rowe, M., Yunusov, T., Haben, S., Holderbaum, W., & Potter, B. (2014). The real-time optimisation of DNO owned storage devices on the LV network for peak reduction. Energies, 7, 3537–3560.CrossRefGoogle Scholar
  18. Sapankevych, N. I., & Sankar, R. (2009). Time series prediction using support vector machines: A survey. IEEE Computational Intelligence Magazine, 4(5), 24–38.CrossRefGoogle Scholar
  19. Schmidt, O., Hawkes, A., Gambhir, A., & Staffell, I. (2014). The future cost of electrical energy storage based on experience rates. Nature Energy, 2, 17110.CrossRefGoogle Scholar
  20. Scottish and Southern Energy Power Distribution: New Thames Valley Vision. (2014). URL http://www.thamesvalleyvision.co.uk/.
  21. Tian, L., & Noore, A. (2004). A novel approach for short-term load forecasting using support vector machines. International Journal of Neural Systems 14(05), 329–335.  https://doi.org/10.1142/s0129065704002078. URL http://www.worldscientific.com/doi/abs/10.1142/S0129065704002078.
  22. Trkay, B. E., & Demren, D. (2011). Electrical load forecasting using support vector machines. In 2011 7th International Conference on Electrical and Electronics Engineering (ELECO). IEEE, Bursa, Turkey.Google Scholar
  23. Vapnik, V. (1995). The nature of statistical learning theory. New York, USA: Springer.CrossRefMATHGoogle Scholar
  24. Yunusov, T., Frame, D., Holderbaum, W., & Potter, B. (2016). The impact of location and type on the performance of low-voltage network connected battery energy storage systems. Applied Energy, 165, 202–213 (2016).  https://doi.org/10.1016/j.apenergy.2015.12.045. URL http://linkinghub.elsevier.com/retrieve/pii/S0306261915016189.
  25. Yunusov, T., Haben, S., Lee, T., Ziel, F., Holderbaum, W., & Potter, B. (2017). Evaluating the effectiveness of storage control in reducing peak demand on low voltage feeders. In Proceedings of the 24th International Conference on Electricity Distribution (CIRED).Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Timur Yunusov
    • 1
  • Georgios Giasemidis
    • 2
  • Stephen Haben
    • 3
  1. 1.TSBE Centre, School of Build EnvironmentUniversity of ReadingReadingUK
  2. 2.CountingLab LTD and CMoHBUniversity of ReadingReadingUK
  3. 3.Mathematical Institute, University of OxfordOxfordUK

Personalised recommendations