Machine learning analysis for a flexibility energy approach towards renewable energy integration with dynamic forecasting of electricity balancing power

One of the most important instruments to be able to provide the needed level of flexibility in the electricity system supporting renewable energy integration are balancing markets. We propose a dynamic approach of balancing procurement using machine learning algorithms. We apply a simulation for a Dynamic Day-Ahead Dimensioning Model to the Austrian delta control area. By using public data on renewables, generation and load we show that dynamic dimensioning and procurement of balancing power enables savings in comparison to static dimensioning and procurement with the same level of security.

[1]  A. Conejo,et al.  Economic Valuation of Reserves in Power Systems With High Penetration of Wind Power , 2009 .

[2]  Mark O'Malley,et al.  Evolution of operating reserve determination in wind power integration studies , 2010, IEEE PES General Meeting.

[3]  Anja Ohsenbrügge.  Dynamische Regel-und Reserveleistungsvorhaltung in zukünftigen Smart Grids , 2015 .

[4]  Oliver Brückl,et al.  Wahrscheinlichkeitstheoretische Bestimmung des Regel- und Reserveleistungsbedarfs in der Elektrizitätswirtschaft , 2006 .

[5]  Lion Hirth,et al.  Balancing power and variable renewables: Three links , 2015 .

[6]  G. Dany.  Power reserve in interconnected systems with high wind power production , 2001, 2001 IEEE Porto Power Tech Proceedings (Cat. No.01EX502).

[7]  Giuseppe Bonaccorso,et al.  Machine Learning Algorithms , 2018, Optimization Techniques and Applications with Examples.

[8]  T. Schluter,et al.  Flexible dimensioning of control reserve for future energy scenarios , 2013, 2013 IEEE Grenoble Conference.

[9]  A. Moser,et al.  Expectation-based reserve capacity dimensioning in power systems with an increasing intermittent feed-in , 2013, 2013 10th International Conference on the European Energy Market (EEM).

[10]  R. Entriken,et al.  Reserve determination for system with large wind generation , 2009, 2009 IEEE Power & Energy Society General Meeting.

[11]  M. Shahidehpour,et al.  Security-Constrained Unit Commitment With Volatile Wind Power Generation , 2008, IEEE Transactions on Power Systems.

[12]  Daniel S. Kirschen,et al.  Estimating the Spinning Reserve Requirements in Systems With Significant Wind Power Generation Penetration , 2009, IEEE Transactions on Power Systems.

[13]  R. Tibshirani,et al.  On the “degrees of freedom” of the lasso , 2007, 0712.0881.

[14]  Gintaras V. Reklaitis,et al.  Operating reserve policies with high wind power penetration , 2011, Comput. Chem. Eng..

[15]  Pedro M. Domingos.  A few useful things to know about machine learning , 2012, Commun. ACM.

[16]  Christoph Maurer,et al.  Dimensioning of secondary and tertiary control reserve by probabilistic methods , 2009 .

[17]  Manuel A. Matos,et al.  Comparison of probabilistic and deterministic approaches for setting operating reserve in systems with high penetration of wind power , 2010 .

[18]  M A Matos,et al.  Setting the Operating Reserve Using Probabilistic Wind Power Forecasts , 2011, IEEE Transactions on Power Systems.

[19]  Christoph Weber,et al.  Impacts of Dynamic Probabilistic Reserve Sizing Techniques on Reserve Requirements and System Costs , 2015, IEEE Transactions on Sustainable Energy.

[20]  2 Machine Learning Algorithms adapted in , .

[21]  Sebastian Lehnhoff,et al.  Dynamic Data Driven Dimensioning of Balancing Power with k-Nearest Neighbors , 2015 .

[22]  R. Tobias.  An Introduction to Partial Least Squares Regression , 1996 .

[23]  W. Winter,et al.  Integration erneuerbarer Energien in die deutsche Stromversorgung im Zeitraum 2015–2020 mit Ausblick auf 2025 , 2011 .

[24]  D. Jost,et al.  A new method for day-ahead sizing of control reserve in Germany under a 100% renewable energy sources scenario , 2015 .

[25]  Sebastian Lehnhoff,et al.  DYNAMIC DIMENSIONING OF BALANCING POWER WITH FLEXIBLE FEATURE SELECTION , 2015 .

[26]  Albert Y. Zomaya,et al.  Toward Energy-Aware Scheduling Using Machine Learning , 2012 .