Kernel selection for power market inference via block successive upper bound minimization

Advanced data analytics are undoubtedly needed to enable the envisioned smart grid functionalities. Towards that goal, modern statistical learning tools are developed for day-ahead electricity market inference. Congestion patterns are modeled as rank-one components in the matrix of spatio-temporal prices. The new kernel-based predictor is regularized by the square root of the nuclear norm of the sought matrix. Such a regularizer not only promotes low-rank solutions, but it also facilitates a systematic kernel selection methodology. The non-convex optimization problem involved is efficiently driven to a stationary point following a block successive upper bound minimization approach. Numerical tests on real high-dimensional market data corroborate the interpretative merits and the computational efficiency of the novel method.

[1]  S. Stoft.  Power System Economics: Designing Markets for Electricity , 2002 .

[2]  N. Amjady,et al.  Energy price forecasting - problems and proposals for such predictions , 2006 .

[3]  V. Koltchinskii,et al.  SPARSITY IN MULTIPLE KERNEL LEARNING , 2010, 1211.2998.

[4]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[5]  L. Zhang,et al.  Energy Clearing Price Prediction and Confidence Interval Estimation with Cascaded Neural Networks , 2002, IEEE Power Engineering Review.

[6]  Zuyi Li,et al.  Market Operations in Electric Power Systems : Forecasting, Scheduling, and Risk Management , 2002 .

[7]  Zhi-Quan Luo,et al.  A Unified Convergence Analysis of Block Successive Minimization Methods for Nonsmooth Optimization , 2012, SIAM J. Optim..

[8]  Marco van Akkeren,et al.  A GARCH forecasting model to predict day-ahead electricity prices , 2005, IEEE Transactions on Power Systems.

[9]  Francis R. Bach,et al.  A New Approach to Collaborative Filtering: Operator Estimation with Spectral Regularization , 2008, J. Mach. Learn. Res..

[10]  Charles A. Micchelli,et al.  When is there a representer theorem? Vector versus matrix regularizers , 2008, J. Mach. Learn. Res..

[11]  A. Ott,et al.  Experience with PJM market operation, system design, and implementation , 2003 .

[12]  L. Tesfatsion,et al.  Short-Term Congestion Forecasting in Wholesale Power Markets , 2011, IEEE Transactions on Power Systems.

[13]  N. Aronszajn.  Theory of Reproducing Kernels. , 1950 .

[14]  Ethem Alpaydin,et al.  Multiple Kernel Learning Algorithms , 2011, J. Mach. Learn. Res..

[15]  Georgios B. Giannakis,et al.  Electricity Market Forecasting via Low-Rank Multi-Kernel Learning , 2013, IEEE Journal of Selected Topics in Signal Processing.

[16]  Georgios B. Giannakis,et al.  Nonparametric Basis Pursuit via Sparse Kernel-Based Learning: A Unifying View with Advances in Blind Methods , 2013, IEEE Signal Processing Magazine.

[17]  M. Shahidehpour,et al.  A Hybrid Model for Day-Ahead Price Forecasting , 2010, IEEE Transactions on Power Systems.

[18]  Georgios B. Giannakis,et al.  Monitoring and Optimization for Power Grids: A Signal Processing Perspective , 2013, IEEE Signal Processing Magazine.

[19]  Charles A. Micchelli,et al.  Learning the Kernel Function via Regularization , 2005, J. Mach. Learn. Res..

[20]  Chen-Ching Liu,et al.  Day-Ahead Electricity Price Forecasting in a Grid Environment , 2007, IEEE Transactions on Power Systems.

[21]  Mohammad Shahidehpour,et al.  Electricity Price Forecasting , 2002 .

[22]  A.M. Gonzalez,et al.  Modeling and forecasting electricity prices with input/output hidden Markov models , 2005, IEEE Transactions on Power Systems.

[23]  J. Contreras,et al.  ARIMA Models to Predict Next-Day Electricity Prices , 2002, IEEE Power Engineering Review.