1. Proceedings
  2. I3M
  3. 2020
  4. EMSS
  5. 10.46354/i3m.2020.emss.045

Trading electricity markets using neural networks

  • a Laura Pozzetti ,
  • b John Cartlidge 
  • a,b Department of Computer Science, University of Bristol, Bristol, UK
Cite as
Pozzetti L., Cartlidge J. (2020). Trading electricity markets using neural networks. Proceedings of the 32nd European Modeling & Simulation Symposium (EMSS 2020), pp. 311-318. DOI: https://doi.org/10.46354/i3m.2020.emss.045

Abstract

We tackle the problem of developing an automated trading strategy to pro t in the British intraday continuous electricity markets. We must train a feedforward neural network to predict one hour ahead total electricity transmission system demand. In live testing to ensure no look-ahead bias, we present forecast results with accuracy better than National Grid’s own demand forecasts. We then train a second feedforward neural network, using our demand forecast as an input to the network, to predict one hour ahead net imbalance volume (NIV), and use this predicted NIV as a trading signal to buy and sell 30-minute electricity contracts. In live testing, between 09 March and 22 March 2020, the trading algorithm made 599 simulated trades, with 431 trades returning a pro t (an accuracy of 72%). These results demonstrate the potential of neural network driven automated trading strategies to make signi cant risk-adjusted excess returns (i.e., pro ts) in the intraday electricity markets.

References

  1. Amira Technologies (2019). Forecast performance report: Autumn 2019. Online https://www.amira-technologies.co.uk/pdf/forecast_report_latest.pdf.
  2. Blaavand, J., Warrington, R., and Leese, R. (2017). Demand forecasting and proposed incentives. Technical report, Smith Institute. Online https://www.ofgem.gov.uk/ofgem-publications/111128.
  3. EPEX (2019). New monthly record on EPEX intraday: Volumes on British, French, and German continuous markets climb to new record high. Online: http://static.epexspot.com/document/41140/2019-11-2005_EPEXSPOT_PowerTradingResults2010-2019.pdf.
  4. Garcia, M. P. and Kirschen, D. S. (2006). Forecasting system imbalance volumes in competitive electricity markets. IEEE Transactions on Power Systems, 21(1):240–248.
  5. National Grid ESO (2020). Platform for energy forecasting (PEF) strategic project roadmap update June 2020. Online https://data.nationalgrideso.com/plans-reports-analysis/platform-for-energy-forecasting-pef-strategic-project-roadmap.
  6. Ofgem (2002). New electricity trading arrangements (NETA) - one year review. Online: https://www.ofgem.gov.uk/ofgem-publications/76199/1109-factsheet110224july.pdf.
  7. Ofgem (2015). Balancing and settlement code (BSC) P305: Electricity balancing significant code review developments. Online: https://www.ofgem.gov.uk/sites/default/files/docs/2015/04/p305d_1.pdf.
  8. Ofgem (2018). Analysis of the first phase of the electricity balancing significant code review. Online https://www.ofgem.gov.uk/system/files/docs/2018/08/analysis_of_the_first_phase_of_the_electricity_balancing_significant_code_review_as_final_version_publication.pdf.
  9. Pozzetti, L. (2020). Algorithmic trading in UK power markets using neural networks. BSc thesis, Department of Computer Science, University of Bristol.
  10. Tribe, E. (2017). P305 post implementation review. ELEXON Online: https://www.elexon.co.uk/wp-content/uploads/2017/02/P305-Post-Implementation-Review.pdf.
  11. Weron, R. (2014). Electricity price forecasting: A review of the state-of-the-art with a look into the future. International Journal of Forecasting, 30(4):1030–1081.
  12. Wilson, G., Godfrey, N., Sharma, S., and Bassett, T. (2020). UK’s COVID-19 restrictions bring about historical lows in energy demand. Channel News Asia. Online: ttps://www.channelnewsasia.com/news/commentary/covid-19-coronavirus-uk-britain-electric
    ity-grid-fuel-demand-12598220.