The energy distance for ensemble and scenario reduction

Scenario reduction techniques are widely applied for solving sophisticated dynamic and stochastic programs, especially in energy and power systems, but are also used in probabilistic forecasting, clustering and estimating generative adversarial networks. We propose a new method for ensemble and scenario reduction based on the energy distance which is a special case of the maximum mean discrepancy. We discuss the choice of energy distance in detail, especially in comparison to the popular Wasserstein distance which is dominating the scenario reduction literature. The energy distance is a metric between probability measures that allows for powerful tests for equality of arbitrary multivariate distributions or independence. Thanks to the latter, it is a suitable candidate for ensemble and scenario reduction problems. The theoretical properties and considered examples indicate clearly that the reduced scenario sets tend to exhibit better statistical properties for the energy distance than a corresponding reduction with respect to the Wasserstein distance. We show applications to a Bernoulli random walk and two real data-based examples for electricity demand profiles and day-ahead electricity prices. This article is part of the theme issue ‘The mathematics of energy systems’.

[1]  G. Pflug,et al.  Multistage Stochastic Optimization , 2014 .

[2]  Gabriel Peyré,et al.  Learning Generative Models with Sinkhorn Divergences , 2017, AISTATS.

[3]  A. F. Adams,et al.  The Survey , 2021, Dyslexia in Higher Education.

[4]  D. Kuhn,et al.  Scenario reduction revisited: fundamental limits and guarantees , 2017, Mathematical Programming.

[5]  Werner Römisch,et al.  Scenario Reduction Techniques in Stochastic Programming , 2009, SAGA.

[6]  M. Wietschel,et al.  The shape matters! How structural changes in the electricity load curve affect optimal investments in generation capacity , 2013, 2013 10th International Conference on the European Energy Market (EEM).

[7]  K. Ball.  Eigenvalues of Euclidean distance matrices , 1992 .

[8]  Xianyang Zhang,et al.  Distance Metrics for Measuring Joint Dependence with Application to Causal Inference , 2017, Journal of the American Statistical Association.

[9]  Carsten Gottschlich,et al.  The Shortlist Method for Fast Computation of the Earth Mover's Distance and Finding Optimal Solutions to Transportation Problems , 2014, PloS one.

[10]  Blazej Olek,et al.  Review of the methods for evaluation of renewable energy sources penetration and ramping used in the Scenario Outlook and Adequacy Forecast 2015. Case study for Poland , 2017 .

[11]  Hans-Peter Kriegel,et al.  Integrating structured biological data by Kernel Maximum Mean Discrepancy , 2006, ISMB.

[12]  Florian Ziel,et al.  Probabilistic mid- and long-term electricity price forecasting , 2017, Renewable and Sustainable Energy Reviews.

[13]  Gábor J. Székely,et al.  The Energy of Data , 2017 .

[14]  Sarah M. Ryan,et al.  Scenario construction and reduction applied to stochastic power generation expansion planning , 2013, Comput. Oper. Res..

[15]  Yiming Yang,et al.  MMD GAN: Towards Deeper Understanding of Moment Matching Network , 2017, NIPS.

[16]  Y. Ni,et al.  An Improved Scenario Reduction Technique and Its Application in Dynamic Economic Dispatch Incorporating Wind Power , 2019, 2019 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia).

[17]  Gabriel Peyré,et al.  Sinkhorn-AutoDiff: Tractable Wasserstein Learning of Generative Models , 2017 .

[18]  Christian P. Robert,et al.  On parameter estimation with the Wasserstein distance , 2017, Information and Inference: A Journal of the IMA.

[19]  Mohammad Shahidehpour,et al.  Security-constrained unit commitment with volatile wind power generation , 2009, 2009 IEEE Power & Energy Society General Meeting.

[20]  C. Robert,et al.  Inference in generative models using the Wasserstein distance , 2017, 1701.05146.

[21]  N. Growe-Kuska,et al.  Scenario reduction and scenario tree construction for power management problems , 2003, 2003 IEEE Bologna Power Tech Conference Proceedings,.

[22]  Ponnuthurai N. Suganthan,et al.  Optimal reactive power dispatch with uncertainties in load demand and renewable energy sources adopting scenario-based approach , 2019, Appl. Soft Comput..

[23]  P. Beraldi,et al.  A clustering approach for scenario tree reduction: an application to a stochastic programming portfolio optimization problem , 2013, TOP.

[24]  Maria L. Rizzo,et al.  Energy distance , 2016 .

[25]  Alain Trouvé,et al.  Interpolating between Optimal Transport and MMD using Sinkhorn Divergences , 2018, AISTATS.

[26]  Qiao Li,et al.  Fast Scenario Reduction for Power Systems by Deep Learning , 2019, ArXiv.

[27]  Pierluigi Siano,et al.  Stochastic optimal scheduling of distributed energy resources with renewables considering economic and environmental aspects , 2018 .

[28]  Maria L. Rizzo,et al.  Energy statistics: A class of statistics based on distances , 2013 .

[29]  Thomas L. Hayden,et al.  Distance matrices and regular figures , 1993 .

[30]  Chan-Nan Lu,et al.  Stochastic Analyses of Electric Vehicle Charging Impacts on Distribution Network , 2014, IEEE Transactions on Power Systems.

[31]  Georg Ch. Pflug,et al.  Multiscale stochastic optimization: modeling aspects and scenario generation , 2019, Comput. Optim. Appl..

[32]  Werner Römisch,et al.  A note on scenario reduction for two-stage stochastic programs , 2007, Oper. Res. Lett..

[33]  Jitka Dupacová,et al.  Scenario reduction in stochastic programming , 2003, Math. Program..

[34]  R. Weron,et al.  Recent advances in electricity price forecasting: A review of probabilistic forecasting , 2016 .

[35]  Vladimiro Miranda,et al.  Impact of clustering-based scenario reduction on the perception of risk in unit commitment problem , 2015, 2015 18th International Conference on Intelligent System Application to Power Systems (ISAP).

[36]  Kenji Fukumizu,et al.  Equivalence of distance-based and RKHS-based statistics in hypothesis testing , 2012, ArXiv.

[37]  Marco Cuturi,et al.  On Wasserstein Two-Sample Testing and Related Families of Nonparametric Tests , 2015, Entropy.

[38]  Benjamin F. Hobbs,et al.  Comparing scenario reduction methods for stochastic transmission planning , 2019 .

[39]  Junpeng Wang,et al.  Visualization and Visual Analysis of Ensemble Data: A Survey , 2019, IEEE Transactions on Visualization and Computer Graphics.

[40]  W. Römisch.  Stability of Stochastic Programming Problems , 2003 .

[41]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[42]  Bernardo K. Pagnoncelli,et al.  Scenario reduction for stochastic programs with Conditional Value-at-Risk , 2018, Math. Program..

[43]  X. Nguyen.  Wasserstein distances for discrete measures and convergence in nonparametric mixture models 1 , 2011 .

[44]  Stephen A. Vavasis,et al.  Local minima for indefinite quadratic knapsack problems , 1992, Math. Program..

[45]  Cornetto.  Test , 2020, ACM Transactions on Algorithms.

[46]  M. Muskulus,et al.  Wasserstein distances in the analysis of time series and dynamical systems , 2011 .

[47]  Nikolaos V. Sahinidis,et al.  Semidefinite relaxations for quadratically constrained quadratic programming: A review and comparisons , 2011, Math. Program..

[48]  Donald Davendra,et al.  A CUDA Approach for Scenario Reduction in Hedging Models , 2018 .

[49]  René Henrion,et al.  Scenario reduction in stochastic programming with respect to discrepancy distances , 2006, Comput. Optim. Appl..

[50]  Juan M. Corchado,et al.  Stochastic interval-based optimal offering model for residential energy management systems by household owners , 2019, International Journal of Electrical Power & Energy Systems.